The Tech Gazette

AI coding agents changed how fast teams ship. They didn’t change who has to verify the output. Last week, Qodo field CTO Shlomo Dalezman and solutions engineer Jonathan Klick ran a live session on what happens when code review stops waiting for the pull request. They walked…

This is a roundup of articles from the openSUSE community listed on planet.opensuse.org. This community blog feed aggregator lists the featured highlights below from Oct. 9 - 15. This week highlights KDE’s 30th anniversary and the premiere of the KDE documentary, the launch of Pl…

This update includes a minor upgrade to Chromium 152 (Extended Stable Release) from upstream, which in turn includes security fixes from Chromium 155.

More on Desktop 8.2 · Desktop

DeepSeek V4.1 Flash 1017 is now available on AI Gateway. DeepSeek V4.1 Flash 1017 is a minor update to DeepSeek V4.1 Flash that improves reasoning, coding, agentic tasks, and image understanding. The coding gains show up in terminal work and in building and editing whole reposito…

More on Flash 1017 · Flash

Reflection’s open-weight Beam model on CoreWeave spans training with Weights & Biases Models, SFT in partnership with Axolotl, and Day 0 API access on Serverless Inference in CoreWeave Forge.

Multi-Statement Transactions are Generally Available, enabling organizations to run...

A new Linux kernel vulnerability, nicknamed PwnEscape ( CVE-2026-99999 ), has been disclosed. It is a guest-to-host virtual machine (VM) escape: anyone who controls a guest VM can send specially crafted network packets that leak the host’s kernel memory, crash the host, or,…

The App Review Guidelines have been revised to support new features, updated policies, and to provide clarification. App Review Guidelines Before You Submit: Added links to additional documentation. 1.2: Clarified app content and app types that are subject to this guideline, and…

Teams running distributed SQL in production face a trade-off. A managed service removes the work of upgrades, scaling, and backups, and it moves the database into the vendor’s cloud account. Self-managing TiDB keeps the database in your account and leaves every upgrade, backup, a…

The App Review Guidelines are a core part of the safe and trusted experience that the App Store strives to offer users around the world — and the foundation of that trust is clear standards that are consistently enforced. As outlined in our App Review Guidelines, pornographic mat…

The OSI has teamed up again with Perforce Zend and OpenLogic, as well as the Eclipse Foundation, for the 2027 State of Open Source Survey.

TidesDB now exists inside MariaDB as a pluggable storage engine. With the TideSQL plugin available, anyone can load TidesDB into a MariaDB server and begin experimenting with it. But having a plugin is only the first step — you still need a practical, reproducible way to run real…

Every turn of the AI loop adds new weights, traces, and access paths. CoreWeave CISO Jim Higgins discusses keeping identity, keys, and telemetry attached to the work as it moves.

General Science

Organizations in regulated industries run container workloads in air-gapped AWS environments, which means migrating hundreds of Helm charts and images into private Amazon ECR with consistent security controls. This post uses Amazon ECR repository creation templates, Terraform, an…

Today, Reflection released Beam, its first open-weight model built to give developers, researchers, and enterprises open control and customization for frontier software engineering and long-horizon agentic tasks. Released under a permissive Apache 2.0 license with BF16, FP8, and…

National Design Studio and Cognition are looking for engineers to help modernize America’s federal software infrastructure.

n8n is easy to try and harder to run under production traffic. A single container with SQLite works great for a quick demo. Production is a different story entirely. You need workers, webhook processors, PostgreSQL, Redis,...

Why Zephon? The data that a model sees during training is one of the most powerful levers we have for influencing how well it performs. Much of the day-to-day work... The post Bringing Elastically Deterministic Data Loading to TorchTitan with Zephon appeared first on PyTorch.

As agents get more capable, the data and evaluations that measure them have to become a science in their own right. Here are the sessions, takeaways and videos from the Frontier Data Summit. The post Frontier Data Summit 2026 Recap appeared first on Snorkel AI.

Fedora Linux 45 is almost here! Developers, packagers, and quality assurance engineers are hard at work putting the remaining polish on Fedora 45, which means it’s time for our Global Virtual Release Party! It is almost time to celebrate! If all goes according to plan, we w…

More on Linux 45 · Linux

Prasanna Vijayanathan, Renzo Sanchez-Silva When Jake Paul and Mike Tyson stepped into the ring in November 2024, we delivered the fight to roughly 65 million concurrent streams. On an ordinary day our infrastructure takes in more than 2 billion requests and handles upwards of 38…

Want to know the latest from Google Cloud? Find it here in one handy location. Check back regularly for our newest updates, announcements, resources, events, learning opportunities, and more. Tip: Not sure where to find what you’re looking for on the Google Cloud blog? Start here…

In production systems today, most AI agents operate as lightweight dispatchers. They parse a prompt, make an API call, retrieve a record, and hand back control. Their execution is fast, stateless and cheap. Biotechnology does not work this way. Folding a protein target, screening…

The 2026 Stack Overflow Developer Survey shows only 7% of developers trust AI output for important decisions. Here’s what that means for running agents safely.

IntroductionSupply chain attacks on package registries have grown in volume and sophistication [1]...

ClickHouse’s new native TCP client for.NET brings faster reads and inserts with fewer allocations, columnar APIs, POCO support, and a path to migrate from HTTP.

How we used Partial Prefetching and a coding agent to improve navigation across v0 with Next.js 16.4.

At Spotify, every surface and interaction is personalized to you as a user: the content on the Home page, the... The post Advancing Generative Recommendation Inference: our collaboration with NVIDIA on TensorRT LLM appeared first on Spotify Engineering.

A recap of every session at O11yDay London 2026, from Sam Newman's cobots and dark factories to Corey Quinn on AI cost observability, plus how Honeycomb debugs its own agents with Canvas, Agent Timeline, and AI Ecosystem.

This update includes a minor upgrade to Chromium 152 (Extended Stable Release) from upstream.

More on Browser 8.2 · Browser

Matrix Live S13E13 - Amandine & Matthew brace for the conference Today's Matrix Live: https://youtube.com/watch?v=w9SWwXVyoPg Dept of The Matrix Conference Andy announces As you can see in Matrix Live this week, we are into the final few days before the Conference!…

Enterprises can now manage their dependencies’ licenses at scale with sophisticated, ruleset-based checks that enforce a centralized policy. Open source license compliance is now generally available for GitHub Enterprise Cloud… The post Open source license compliance…

One RSS reader, on every device you own, no algorithms invited.

Junie Local is supported on the NVIDIA RTX Spark from day one, and Windows support moves from experimental to stable. NVIDIA’s RTX Spark platform pairs a Blackwell RTX GPU with a Grace CPU to run larger local AI models in slim laptops and compact desktops. Junie Local leverages t…

Beam is now available on Baseten. Reflection’s 501B open-weight MoE model is built for efficient reasoning in coding and agentic workflows, with 1M context.

A single account on its own tells you very little. But look at who it sends money to, who those accounts send money to, and suddenly you have got a lot to work with. That’s the idea behind graph embeddings:… Read more →

Identity alone doesn't guarantee a bot will behave responsibly. We are introducing a system to continuously evaluate whether Verified bots honor site preferences, like robots.txt and HTTP 429s, ensuring they remain accountable over time.

What’s the most vital asset for an AI company today?  For venture backed AI business, it might be “funding” or “revenue”. But, there’s another asset that’s just as critical: trust.  At Mozilla, building technology that people and businesses can genuinely trust has…

We’ve recently been dealing with an interesting customer case where a PXC version upgrade from 5.7 to 8.0 seemed to trigger a completely unexpected problem. Even though the workload and MySQL settings did not change, all PXC members began exhibiting sparse but regular bursts of m…

PyPI improves on checks for the reputation of an email address when used to register an account or add to an existing account, and rejects addresses identified as disposable. With the new check in place, outlook.com and hotmail.com addresses can once again be used on PyPI. Backgr…

Feature · gpt-live-1 · v1/audio/voices · v1/live/sessions. Released prompt-to-voice for all API customers. Create a custom voice from a text description through the API or voice Playground, then reuse it in GPT-Live sessions. Creating a prompt-generated voice is free; speech gene…

Stripe Data Pipeline now supports real-time sync to Postgres. Connect Neon from the Stripe Dashboard and query live payment, subscription, and customer data next to your app data, without building API or webhook integrations.

How we made VM Sandbox filesystems fast by speaking in blocks instead of files.

In this blog, I will share my Golden Kubestronaut journey, how I became the first and youngest Golden Kubestronaut from Nepal, what motivated me to pursue this path, and what it actually takes to earn the...

Find out the difference between On-Behalf-Of and OAuth Token Exchange, and why the two are often confused.

What happened when we tried to hand 60 microservices a major framework upgrade, and why we stopped expecting AI to figure things out on its own. Sixty services, one chassis We’re the team behind Doctoboot, Doctolib’s internal Spring Boot chassis. A chassis is shared plumbing: the…

As an Android developer, you’ll almost always have access to a device. However, when handling hardware, the device that you actually need often happens to be already in use, set up incorrectly, stored in another lab, or simply waiting to be reset. That’s when Android development…

Enterprise teams want to build apps connected to their company’s systems and data. But strict security requirements often mean development and production must stay within the isolation and control of their own cloud. Replit has steadily expanded its governance and isolation postu…

How we moved Logfire's million-minute CI workload to RunsOn and used ARM, Spot, caching, and right-sized EC2 instances to cut our monthly CI bill by roughly two-thirds.

“A benchmarking case study on relation extension locks, partitioning, and peak write load”; I gave this talk at PG Pune 2026, a community Postgres event with about 170 attendees, mostly database engineers, DBAs and Postgres enthusiasts. This post is the written version, plus two…

Ryan is joined by Gabriela Moreira, CEO at Quint, about why formal specification languages like TLA+ are essential for catching system design flaws before writing code, how Quint brings developer-friendly syntax and executability to formal methods, and the ways trace validation a…

Agents already shop on the web, but they often use pages and tools originally built for people and search engines. Building a better shopping experience for agents means treating them as first-class shoppers, with product information they can read directly and tools designed for…

Take a look around the unique Founders, Inc. campus incubating what hope to be future-defining technology companies. The post Building the next generation of hardware founders appeared first on Raspberry Pi.

This October I attended both the openSUSE.Asia Summit 2026 in Yogyakarta, Indonesia, and Open Source Summit Europe 2026 in Prague, practically back to back. Two conferences, two continents and one very long trip. Getting to both was quite an adventure. The way there was a train f…

LeRobot trains directly from Lance datasets in object storage, a hybrid scan mines robot fleet data for edge cases, and distributed search reaches 1,066 QPS across 10B vectors, plus upcoming events, enterprise updates, and community contributions.

Lance Blob V2 introduces adaptive storage semantics, easily upload Lance datasets to Hugging Face Hub, and OpenClaw establishes LanceDB as a default memory layer for agents, plus community and enterprise updates.

LanceDB lets you tune search precision at query time, balance memory and disk caching, and keep new data searchable as your table grows.

You can now use the generally available Xcode 27 runner image to build and test your Apple applications in GitHub Actions. Previously in public preview, this GitHub-hosted image runs on… The post Xcode 27 runner image generally available appeared first on The GitHub Blog.

More on Xcode 27 · Xcode

@GitHub in Slack and Teams now supports working across multiple repositories in a single conversation. Whether you’re creating issues, opening pull requests, or investigating bugs, you can work across repositories… The post Multi-repository support and other improveme…

Explore September's models, Routines, Tool Search, Model Router, and Insights through one two-agent workflow that turns public releases into a Foundry developer changelog. The post What’s new in Microsoft Foundry | September 2026 appeared first on Microsoft Foundry Blog.

More on Pretraining 2.8 · Pretraining

Z.ai’s GLM-5.3 is the top open-weight model on Epoch’s Cyber ECI, scoring where the closed-weight frontier stood 6–7 months earlier.

Reflection's first open-weight model is now available on Ollama, and free for a limited time only. Beam is designed for software engineering, reasoning, and long-running agentic tasks. Access Beam through Ollama Cloud.

Glimpse from Elorian is now available on AI Gateway. The visual reasoning model accepts images and video to follow objects through motion, interpret spatial relationships, and connect an action to its physical consequences. Glimpse can return to visual evidence as it works: zoomi…

Perplexity Decider V1.2 27B and Decider Flash V1.2 4B are now available on AI Gateway. Both are multimodal decision models for classifying content, routing requests, and scoring text, structured records, and images. The models answer typed questions with probabilities. Boolean qu…

Beam from Reflection AI, previously available in stealth as stealth/cat-jam, is now available on AI Gateway. Beam is free to use for a limited time during the launch period when serviced by Reflection AI. Beam is an low latency US open-weight reasoning model for coding and agenti…

Enterprise teams can now create, track, and reply to support tickets without leaving Netlify in our new Support Hub. Go to User settings > Support to see your tickets, check their status, and continue your conversation with the Netlify support team. Team Owners can see every tick…

TypeSafe’s Jev 1.14.0 model is now available through Netlify’s AI Gateway with zero configuration required. Install @typesafe-ai/sdk and use it directly in your Netlify Functions without managing API keys or authentication. AI Gateway handles credentials automatically, and usage…

More on Jev 1.14.0 · Jev

We’re introducing three updates to help fintechs launch products and grow internationally more easily: simpler payout management with Stripe Global Payouts, support for stablecoin cards in more countries, and unified APIs for building global financial products.

Cloud agents can now build and share interactive canvases. The Environments API is available in beta, environments support up to 300 repos, and large environments load 5x faster.

After 6 years at PostHog, I am delighted to share that I have learned exactly 6 things. 80% of being a good manager is good hiring The other 20% is…

JSON often arrives as a text column inside a table, not as a file. In this post, we infer a schema from such a column, adjust it, and turn the column into typed columns, without first writing the data to a file.

We put NVIDIA's new Vera CPU through its paces using the full TPC-H workload on scale factor 1,000.

As AI-powered threats grow more complex, traditional security operations struggle to keep pace. Google Distributed Cloud AI SecOps, powered by Elastic, brings AI-driven detection, investigation, and response to fully air-gapped environments.

Police reform depends on making data work across force boundaries. Connecting data where it lives gives policing national reach, keeps local accountability, and makes AI cheaper and faster to scale.

Introducing interactive refactoring in gopls and the LSP: a new capability that brings dynamic, multi-step refactorings into your favorite editor.

The following JEP is targeted to JDK 28: 545: Faster Startup and Warmup with ZGC

How agent-assisted CuTe DSL optimizations accelerate Kimi K3 decode on NVIDIA GB300, and how scoped changes, correctness tests, and reproducible benchmarks help ship them responsibly.

Inside the Kimi K3 optimization effort: understanding the bottlenecks and the decisions that made inference faster

vLLM supports Reflection's Beam-1.0 (501B total, 23B active) from day 0, with agent-driven optimization that cut decode latency 2.71× and DFlash2 speculative decoding with adaptive verification.

In this blog you will learn how to write a fast FP16 GEMM kernel for AMD RDNA™ 3 and RDNA™ 4 GPUs, one optimization at a time. You will see which techniques pay off on each architecture, why some help one and hurt the other, and how to measure the effect of every change in time,…

You can now see and control your macOS or Linux runner's desktop. This is nice for trying native apps that Amp builds for you, such as iOS apps: Turn on Share This Mac's Display under Settings › Runner in the macOS app, or start your runner with amp --no-tui --desktop. Yo…

AI Search keyword search with the trigram tokenizer now matches Japanese, Chinese, and Korean queries, including full natural-language questions such as 経費精算の締め日はいつですか？. Previously, a query without spaces was treated as a single exact phrase and usually returned no keyword result…

Agents can turn ideas into pull requests in almost no time. The hard part now is deciding what to build and validating that generated code is safe to ship. Here is a risk-based approach to both.

As previously announced, support for Istio 1.29 has now officially ended. At this point we will no longer back-port fixes for security issues and critical bugs to 1.29. We highly recommend that you upgrade to the latest version of Istio (1.31.2) if you haven’t already.

More on Istio 1.29 · Istio

Hi, it’s Vipul. Let’s explore this week’s changes in the Rails codebase.

Highlights Several core rules have been updated to report more linting problems. no-use-before-define The no-use-before-define rule has been updated to account for the fact that IIFEs execute immediately by treating them as part of the execution context where they are defined and…

More on ESLint v10.13.0 · ESLint

How one business development representative clears his inbound queue and drafts bilingual customer emails with Claude Managed Agents and Claude Tag.

This change has been a long time coming, and I’m excited to tell you about it. The Marketing SIG is now the Promo SIG! If you haven’t run into us before, the Marketing SIG is one of the many SIGs (Special Interest Groups) that make the AlmaLinux project do all the coo…

Today KDE releases a bugfix update to KDE Plasma 6, versioned 6.6.7. Plasma 6.6 was released in February 2026 with many feature refinements and new modules to complete the desktop experience. This release adds three months’ worth of new translations and fixes from KDE’s contribut…

More on Plasma 6.6.7 · Plasma

The Claude SDKs now emit OpenTelemetry traces natively. Five tips to trace your agent with standard OpenTelemetry, see each run as one trace, find where the tokens go, score every run, and compare the outcomes from different prompts.

Better guesses about available capacity meant fewer failed requests. Our jobs spent less time waiting, and we avoided costly moves to other regions.

Why executive judgment decides who wins with AI

See how the team behind Foxy Dumplings built one Unity project for PC, console, and mobile, covering platform abstraction, input, profiling, and Cloud Save.

As Timor-Leste deepens its integration with ASEAN and the global digital economy, continued progress in technical capability, resilience, and Internet governance will be critical to ensuring connectivity delivers broader economic and social benefits.

When you’re on the go, your phone is often the only observability tool you have. And what you can do with it in the first few minutes matters. With the incredible innovation that has gone into mobile CPUs, GPUs and displays, a phone is a powerful device where you can immediately…

At Observability Summit Europe 2026 in Prague, I presented a keynote titled Carbon as a First-Class Signal: Designing a Standard for OpenTelemetry. The talk explored how Software Carbon Intensity (SCI) could become part of cloud-native observability...

At Databricks, we continuously develop and evaluate new AI research techniques to...

If your business runs on Stripe, your application needs access to Stripe data at...

The challenge: Two decades of logging noise at global scaleAtlassian is a global...

More on Switch 4.1 · Switch

Earlier this year, we launched AI Inbox in Gmail which helps users catch up on what matters most. Starting today, we’re expanding AI Inbox to paid Google Workspace users. By highlighting key actionable to-do’s and summarizing priority work updates across your emails, AI Inbox mak…

You can now download a CSV report of your GitHub Code Quality license usage to reconcile your license count with your own records. The report includes: Each active committer who… The post Download GitHub Code Quality license usage as a CSV appeared first on The GitHub Blog.

We are pleased to announce the availability of the release candidate for Microsoft C++ (MSVC) Build Tools version 14.52. With this release, C++23 is ready for production use and available through the /std:c++23 compiler option; see below for details. This is the last MSVC v14.52…

Alibaba DAMO Academy Chief Scientist Dr. Jianyi Meng on what it took to achieve an on-chip AI milestone for RISC-V, and his hopes for RISC-V Summit China 2026. The post <span class="vq-prefix">V Questions with… </span>DAMO Academy’s Dr. Jianyi Meng app…

AI, agents, voice, and physical controls are shifting software development from operating tools to expressing intent—and helping developers stay focused on building. The post The next developer interface is ambient intelligence appeared first on The GitHub Blog.

At OCP Global 2026, Arm AGI CPU and a growing partner ecosystem show how open infrastructure can bring CPUs, accelerators, memory, software and rack design together for scale The post Arm at the OCP Global Summit 2026: How Arm AGI CPU is scaling agentic AI infrastructure from CPU…

We will deprecate the following models across all GitHub Copilot experiences (including Copilot Chat, inline edits, ask and agent modes, and code completions) on November 16th, 2026: Model Deprecation date… The post Upcoming deprecation of selected GitHub Copilot models in…

Eternal (Zomato, Blinkit) cut logging costs 60% and benchmarking costs 80%, and simplified ops by migrating from Apache Kafka® to WarpStream.

More on Logs 25 · Logs

Character.AI streams 5 GiB/s on WarpStream at ~90% lower cost. The lead engineer on why the best infrastructure is gloriously boring.

More on Runs 5 · Runs

Meta’s newest app, Forum, is for anyone who wants to go deeper into the conversations in their Facebook groups. To build Forum, Meta engineers adopted a whole new way of working by bringing an AI agent into the development cycle. We’re sharing insights into what it to…

Every four years, the World Cup becomes the largest planned stress test of our infrastructure. This year it also set an all-time record! There are a few moments throughout the year, like New Years Eve, that really let us see the collective impact of everything we’ve built — but n…

Learn what utility companies should consider before deploying AI, and see how North can help meet the sector's security, data, and operational needs.

Learn how to set up Amazon Redshift Data Sharing between a producer and consumer cluster and use dbt Cloud to transform the shared data into analytics-ready models, with live, zero-copy access and no ETL between clusters.

Put native GitHub features to work to review outside collaborator access and archive stale repositories. The post Staying Evergreen: How we automated secure access and repo hygiene in a shared org appeared first on The GitHub Blog.

Today, Meta is releasing the weights for Muse Spark Open, the most capable US open-weights model to date. It’s available today on Hugging Face under the Muse Model License. Muse... The post Muse Spark Is Now an Open-Weights Model appeared first on PyTorch.

Spatial SQL and spatial joins now run on Lakehouse//RT and they're up to 38X faster...

Muse Spark Open is now available on Baseten. This 816B open-weight omnimodal model is optimized for coding agent workflows, with a 1M token context.

Anthropic announces updates to embedding independent evaluators

Cloudflare Gateway and Data Loss Prevention have inherited the flexibility and power of Cloudflare Workers through our full-stack integration with the Cloudflare Developer platform.

What we shipped at Interrupt NYC and London: LangSmith AI Search, Engine v2, Trajectories, Custom Apps, Fine-Tuning, and Managed Deep Agents v0.9.

Runtime configuration has been brought to the Front-End Web Cloud Native Buildpack in a way that is standard, secure, and requires zero re-builds. By revisiting Twelve-Factor principles through the lens of modern web standards, we have created a strategy that works for any framew…

A new beginning for Windows PCs starts tomorrow with the launch of NVIDIA RTX Spark — one PC built for creators, gamers and AI agents. All are invited to celebrate with NVIDIA and partners at launch events around the world. Share Your Spark Every PC has a story. What was your fir…

Client-side encryption (CSE) gives organizations ultimate control over who can access their most confidential data, using customer-managed encryption keys that render the data unreadable to unauthorized third parties, including Google. Google Drive now supports previewing CSE fil…

Learn how the new dotnetup tool helps macOS and Linux developers install, manage, and update user-local.NET SDKs and runtimes. The post Introducing dotnetup: A toolchain manager for.NET appeared first on.NET Blog.

How to roll out software factories across an engineering org: start with chore automations, meet engineers in their tools, and track factory adoption.

Accelerating the search for novel antimicrobial peptides using Google Antigravity for Science.

Ke Mao, Mark Harman, and Yue Jia have been recognized with the ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA) 2026 Most Influential Paper Award for their 2016 research, “Sapienz: multi-objective automated testing for Android applications.” The recogn…

Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security. The post Proof-based genera…

For a decade, Vercel has shaped how the web is built. Now, we’re doing the same for agents. The companies that win the next decade will build on infrastructure designed for agents from the start, and thousands of people gathered across London, Berlin, New York, Sydney, and San Fr…

October 12 - October 16 Looker Data Apps are now in Preview to build and host interactive, agent-generated applications natively inside Looker Generated using natural language prompts, the Antigravity CLI, or Claude Code, these interactive web applications, natively hosted inside…

It all started with a question: What does Firefox sound like to you? Over the past year, Mozilla developed a small but meaningful set of sounds to add a sonic dimension to the Firefox experience. When we started this process, we knew we wanted something warm, human, and represent…

Dynata reimagined its development lifecycle with AI, moving from AI-assisted coding to the AI-Driven Development Lifecycle (AI-DLC) methodology with Kiro IDE. Learn how the company reduced the development phase of full-stack features from three to four sprints to roughly one, wit…

At Percona Live Europe 2026, Dmitry Lenev and I presented JavaScript in My Database? Over My Dead Query, the story of component_js_lang, the V8-based JavaScript engine for Percona Server for MySQL 8.4. That experience told us few things: The presentation was seen as “for develope…

The first half of October belonged to the ecosystem. The MariaDB Ecosystem Hub passed 200 entries six months after launch, Laravel 13 gained native MariaDB Vector support through community contributors, and Frappe and Solanica joined the Foundation as Silver Sponsors. The Foundat…

AI agents are increasingly trusted with long, open-ended work. They fix bugs across large repositories, operate terminals for hours, and chain numerous tool calls toward a single goal. The strongest models now solve a meaningful share of such tasks, yet on the hardest ones failur…

Code scanning AI Scan for pull requests is now generally available. It complements CodeQL by finding vulnerabilities in pull request changes for languages and frameworks that CodeQL does not currently… The post AI Scan for pull requests generally available appeared first on…

Google Summer of Code 2026 has wrapped up! This was the Kotlin Foundation’s fourth year participating in GSoC, bringing new contributors into the Kotlin open-source ecosystem through hands-on projects and mentorship. This year, two projects successfully reached completion, tackli…

NVIDIA today released Nemotron 3.5 Super VL, an open 120-billion-parameter model that delivers the fastest task completion and highest accuracy for specialized always-on agents. Joining it are two new edge models: Nemotron 3.5 Bolt 74B-A9B 24GB, optimized to balance intelligence…

Code-based mocks are mock servers written in JavaScript. They remember state, fail on demand, and your agent can create, run, and deploy them from the terminal. The post Introducing Code-Based Mocks in Postman appeared first on Postman Blog.

Frontier image quality with the customization and speed of open weights: 5.6B parameters, first in text-to-image, multi-reference generation and editing.

Timeline: FLUX 3 → Model sheet →

More on FLUX 3 · FLUX

Nobody approves a project called “add complexity.” It builds up one shard, one region, and one compliance requirement at a time, until a single database has turned into a fleet and the team running it has grown to match. Atlassian reached that point with Forge SQL, its hosted rel…

Offload settled Postgres data to ClickHouse to reduce storage and query costs while keeping active transactions in Postgres.

AI agent permissions determine what an agent is allowed to do on someone’s behalf. As agents get more powerful, we’re running into a basic problem: a model can’t reliably verify those permissions just because someone claims to have them in a conversation. The problem agents have…

Two LF Decentralized Trust (LFDT) communities—the ToIP / DIF Decentralized Trust Graph Working Group (DTGWG) and OpenVTC, an LFDT lab—reached a major milestone at the Linux Plumbers Conference and LF Open Source Summit EU in Prague last week: demonstrating the first verifiable tr…

A third of employees admit to having fed proprietary, confidential, or sensitive company information into a unauthorized public AI service. The post Two-thirds of business leaders believe their company has already suffered an AI data breach appeared first on WRITER.

Discover how to calculate the true total cost of ownership (TCO) for your data stack by accounting for infrastructure, engineering labor, and transition planning.

Stop solution builds after the first project failure so you can return to fixing the error sooner. The post Stop waiting on a failed build appeared first on Visual Studio Blog.

Introducing Void: an open-source SDK, CLI, and self-hosted platform for building full-stack Vite apps and deploying them to Cloudflare.

Claude Code writes better Cypress tests when it knows your conventions, can check its work in a live session, and can see your run history. Learn how to set up Cypress AI Skills, Cloud MCP, and cypress tap, and which pieces your team needs.

Adaptive Metrics, which is part of the Adaptive Telemetry suite, looks at how your metrics are actually used and recommends how much detail you should keep for each metric. This helps you cut costs while still retaining the data you actually need. Last year we added auto-apply ca…

Agentic autofix for code scanning alerts is now generally available. It remediates alerts by working across your codebase the way a developer would: it explores relevant files, proposes a fix,… The post Agentic autofix for code scanning alerts general availability appeared…

As of October 15, 2026, email notifications for repository push events have been retired and are no longer sent. We shut down this feature to help prevent spam and abuse.… The post Email notifications for repository push events have been retired appeared first on The GitHub…

Over a typical 2 day window, we record 10 times as many emulator sessions lasting more than 30 minutes for vNext as for the legacy emulator. Our bet is that agentic development is helping drive the difference! The appeal is straightforward: coding agents need a database they can…

Dial in the next gaming session this GFN Thursday. New NVIDIA DLSS 4.5 controls give GeForce NOW members more choice over how supported games look and perform in the cloud, even without the latest hardware. Then, step into the Shroud as Keen Games’ Enshrouded leaves early access…

More on DLSS 4.5 · DLSS

Photos hold some of our most meaningful memories: the people we love, the milestones we celebrate, and the everyday moments we want to remember. They connect us to our past and help us share our stories. The post Meet the new Photos app on Windows 11 appeared first on Windows Exp…

More on Windows 11 · Windows

Every day, your PC helps you get things done, stay connected, capture ideas and make time for what you enjoy. That’s why Windows 11 continues to evolve, starting with a faster, more dependable and refined experience. Thro The post What’s new in Windows 11: Everyday experiences, r…

More on Windows 11 · Windows

A new PC is a meaningful purchase. Whether you are upgrading a family laptop, shopping for a student, or someone who loves to create or game, you want your money to go further. The right PC should deliver value when you buy The post Incredible deals and more to love in a new Wind…

How Cloudflare and OpenAI are working together to help AI find useful content, keep up with changes, and make fewer wasted requests — making the experience better for users and website owners alike.

Universal Cart helps make holiday shopping easier. It’s now available in the U.S. across Search and the Gemini app.

The latest interim release of Ubuntu previews privacy conscious speech-to-text, enables post-quantum package signing, completes the rollout of Rust coreutils, and improves enterprise identity integration. October 15, 2026 Today Canonical announced the release of Ubuntu 26.10, cod…

More on Ubuntu 26.10 · Ubuntu

Vector stores have been part of the Python experience in Semantic Kernel for some time. With Microsoft Agent Framework, we have ported that work forward rather than starting over. This post focuses on Python. For the.NET equivalent, see Use vector stores in.NET AI apps. The found…

State-of-the-art probabilistic forecasting in four sizes (5M–209M params), CPU-efficient, zero-shot ready.

Groq is the premier neocloud for fast inference. One fully integrated platform for infrastructure, inference, and control. Millions of developers run trillions of tokens on Groq every week.

Feature · gpt-live-1 · gpt-6-luna · v1/live/sessions. Added native Decisions integration in GPT-Live. Your app can act while your user is speaking or react to what Live says. Let Live call Decisions directly through native delegation, or trigger decisions as input or output trans…

The Compliance API now returns the text each organization member saved as Instructions for Claude in claude.ai and as Global instructions in Cowork, in beta for Claude Enterprise organizations. Call GET /v1/compliance/organizations/{org_uuid}/user_preferences with your existing C…

Neon Realtime is live in public beta. Any SQL query can now be a live query. Subscribe to it and the result stays up to date as your data changes. That makes your whole app reactive, with no polling or refetch logic.

Pinecone BYOC now supports on-demand capacity mode, so you can create indexes in your own cloud account without sizing shards upfront.

GPT-Live, the latest Claude models, Gemini 4 Argon, Gemini Omni, Lyria, Perplexity, Firecrawl, Parallel and Jev are now in Lovable. Describe what you want to build and add voice, reasoning, video, music and live web research to your app.

Deploy on Azure is now in private preview, so your apps and their data can run in your own Azure subscription. Here's every way to build with Lovable on Microsoft today, from fully inside your Microsoft environment to fully managed by Lovable.

How financial services marketing teams can assess the risk of an app, control who sees which data, and check what matters before publishing with Lovable.

ExtractBench Pro is an extension to ExtractBench, derived from real long-tail documents. It targets six new challenges, inspired by the ways in which real customers have broken our assumptions.

It’s that time of the release cycle once again: the beta snapshot! With this, 4.8 is fully in feature-freeze, and developer resources have shifted exclusively to creating and reviewing various regression fixes. For those interested in aiding us on our quest to squash any bugs tha…

More on Godot 4.8 · Godot

Over the past three weeks, our team of Forward-Deployed Researchers (FDRs) post-trained NVIDIA’s Nemotron 3.5 Super VL in partnership with Gong, Deel, and Alice to deliver specialized models that surpass or match frontier quality on their defined use case – yet at a fraction of t…

Also reported by vLLM · Baseten · What's New with AWS

Timeline: Nemotron 3.5 → Model sheet →

More on Nemotron 3.5 · Nemotron

Let agents decide what to build next: self-improving software factories that proactively fix bugs, refine their own skills, and mine user feedback for roadmap work.

Nirmata created Kyverno and donated it to the CNCF in 2020. Kyverno is now a graduated CNCF project and is widely used for enforcing and automating configuration security and best practices.  However, admission control answers one...

Much of today’s scientific progress begins with computation. But turning accelerated computing’s potential into scientific progress — modeling climate systems, designing molecules, simulating materials and more — depends on more than breakthrough technology. Especially for smalle…

Netflix plugged JetBrains’ headless code-intelligence engine – the same one behind IntelliJ IDEA – into Claude Code and Codex, then benchmarked it against a variety of scenarios. The details and numbers are available below. At a glance Company Netflix – Application Developm…

MPS 2026.2 is almost ready! Today we’re making a release candidate build available so you can try out the upcoming release ahead of its official launch. Download MPS 2026.2 Release Candidate now: DOWNLOAD MPS 2026.2 Release Candidate Along with numerous bug fixes, the RC brings s…

More on MPS 2026.2 · MPS

Where to start with agent evals, offline vs online evals, evaluator types, LLM judges, benchmarks with Harbor, multi-turn evals, and keeping costs down.

Long-running agents degrade as context piles up. Learn how offloading, summarization, skills, subagents, and memory keep your agent's context window focused.

The KotlinConf 2027 Call for Papers is now open! We invite developers from across the Kotlin community to bring their ideas, experiences, and expertise to the stage in Kraków. Do you have a Kotlin lesson or discovery that could help other developers? Perhaps you solved a difficul…

Cohere proposes acquiring AISLE to enhance sovereign AI security for enterprises and governments.

We are pleased to welcome Solanica, the company driving the development of OpenEverest, as a MariaDB Foundation Silver Sponsor. OpenEverest is an open-source platform for deploying and managing databases on Kubernetes. Its recently added MariaDB support gives developers and platf…

AI Gateway Production Index — October 2026 Every month, AI Gateway routes tens of trillions of tokens between production applications and AI labs. That traffic gives us a view of what AI usage actually looks like in today's enterprise, and we publish it here monthly. See the Prod…

AI adoption in the enterprise is growing rapidly, and expanding beyond engineering users. In the last three months, token usage by enterprise companies on AI Gateway has quadrupled, and OpenAI recently reported that from August 2025 to June 2026, non-developer Codex users grew 18…

Agents often handle complex work that continues across runs. As that work evolves, they may need a new environment with different tools, runtimes, or compute. They should be able to continue working with the same files when they change environments. We built Vercel Drives to keep…

Last year, we expanded zero-config support for Python and TypeScript backend frameworks. By this summer, developers could deploy their frontend and backend together with Vercel Services and coordinate background work with queues and durable workflows. Today at Vercel Ship in San…

Python on Vercel is now generally available. Build and run Python applications with the frameworks, libraries, and tools you already use. Vercel automatically detects supported frameworks and configures the build and deployment. Your applications run on Fluid compute, which manag…

Vercel Token Pass is now in early access for Pro teams. For $30 per month, your team gets $80 in monthly credit for GLM 5.3 and Kimi K3, with more models to come. Use Token Pass with coding agents such as fx, Command Code, and Pi, or build your own personal agent with the AI SDK.…

Vercel Daemons is now in public beta, allowing you to run long-running processes that pull and execute work from Vercel Queues, databases, or third-party systems without requiring an HTTP server: Consumers using Vercel Queues poll mode, Celery, BullMQ, RabbitMQ, or Kafka. Tempora…

What runs when you create a branch: which settings are copied, where your migrations and functions come from, and what order the steps run in.

Sync real-time Stripe data to Supabase from the Stripe Dashboard. Data Pipeline backfills and keeps it in sync, with no webhooks or rate limits to manage.

Type-safe, composable middleware for Web Fetch handlers. Write it once, use it on any fetch runtime or inside a framework.

Six of the vendors behind openDesk, including Element, have formed Open Productivity: a joint venture that brings the sovereign digital workplace to businesses and critical infrastructure operators across.

The Nordic countries have become the European leader in unicorn startup density, producing the most emerging companies valued at $1 billion or more per capita on the continent. Now, a new chapter for the region is unfolding as countries race to build one of Europe’s most conseque…

Building off the Config momentum in India, Figma is launching an education program, new research, and an MCP partnership with Sarvam.

Here’s a look at everything we launched today to bring teams and agents, and the design, code, and context they work with, into Figma’s full-stack creative canvas.

Today, code layers moves into open beta. Now teams can explore and build prototypes backed by code on the canvas.

We’re launching Specs, a container for your context that lives on the Figma canvas.

Multimodal agents commonly generate freeform reasoning before each action. For small models, limited model capacity can result in lengthy reasoning that provides little useful guidance for action generation while incurring substantial inference cost. To address this challenge, we…

Tool-augmented LLM agents discover relevant APIs from large registries before planning execution, yet finite context windows prohibit injecting all tool definitions into a single prompt. The dominant approach, which we term Retrieve-then-Plan (RtP), embeds the user query and retr…

Meta Muse Spark Open, Meta's open-weight model for agentic coding, is live on Together AI with a 1M-token context window on serverless and dedicated infrastructure.

Weaviate v1.39.3 fixes a medium-severity authorization issue in the nearObject GraphQL operator.

d1-ultra, the world's best decision model for text and images, built for hard real-world cases. Available on the Liquid AI API

Google OAuth Platform now supports DPoP (Demonstrating Proof-of-Possession) to sender-constrain refresh tokens, neutralizing Infostealer malware and adversary-in-the-middle attacks. By cryptographically binding long-lived refresh tokens while maintaining short-lived access tokens…

The google/adk-recipes repository unifies Agent Development Kit (ADK) resources into three structured layers—core/, plugins/, and contrib/—serving as a single source of truth for building AI agents. Developers can leverage reference patterns, install modular vertical skills (such…

A practical engineering guide and reference architecture for establishing continuous evaluation pipelines for multi-turn AI agents using Gemini Enterprise Agent Platform, Google ADK, and OpenInference standards.

d1-ultra from Liquid AI is now available on AI Gateway. It is a multimodal decision model for classification, routing, and scoring across text and images, including visual defect inspection and checking damage claims against customer photos. Rather than generating text, d1-ultra…

Muse Spark Open from Meta is now available on AI Gateway. The open-weight reasoning model is designed for coding and agentic workflows, including multi-file refactoring, debugging, and understanding large codebases. Muse Spark Open supports text and image input with a 1M-token co…

Vercel Connect now supports Secure Compute, allowing connectors to reach providers with IP allowlists, such as Salesforce and Snowflake. With Secure Compute routing enabled, Connect’s provider requests use the selected network’s dedicated static IP addresses instead of Connect’s…

Cat Jam is now available on AI Gateway as a stealth model, free to use during the stealth period. Cat Jam is a reasoning model for coding and agentic tasks, with low-latency inference. It supports function calling, so agents can connect it to your own tools, and structured output…

Vercel Drives is now generally available for Vercel Sandbox on Hobby, Pro, and Enterprise. A Drive is persistent storage that you mount as a directory in a Vercel Sandbox. Its files remain available after the sandbox stops, so later runs and other sandboxes can access them by mou…

Enterprise teams can now run Vercel Functions in their own AWS account with Bring Your Own Cloud (BYOC). With BYOC, your function code and execution stay inside your security and compliance boundary. Vercel continues to handle deployments, environments, routing, and static assets…

Vercel Connect now supports Enterprise-Managed Authorization (EMA) in public beta. Your apps and agents can access supported services on behalf of signed-in users without separate consent prompts for each service. Here's how access works: Admin-managed access: Administrators can…

Vercel Connect now supports MCP Events in public beta, allowing your apps and agents to subscribe to webhook events from compatible MCP servers. Build agents for tasks like summarizing new messages, reviewing newly created issues, or responding to document feedback. Connect creat…

Python is now a first-class language and runtime on Vercel. Build APIs, full-stack applications, and AI agents with native Python SDKs and a shared development and deployment workflow for your Python backend and JavaScript frontend. Python framework and runtime support Vercel han…

Jev 1.14 from TypeSafe AI is now available on AI Gateway. Jev is a decision model for software. You pass it state and a set of questions, and it returns typed Choice, Score, and Boolean answers with probabilities instead of free-form text. It answers every question in one request…

More on Jev 1.14 · Jev

The PostgreSQL Global Development Group announces that the first release candidate of PostgreSQL 19 is now available for download. As a release candidate, PostgreSQL 19 RC 1 will be mostly identical to the initial release of PostgreSQL 19, though some more fixes may be applied pr…

More on PostgreSQL 19 · PostgreSQL · RC 1 · RC

Applications rely on current Stripe data to fulfill orders, update customer accounts, and manage access to paid features. Until now, developers have had to build and maintain custom integrations to keep application databases in sync with Stripe. We’re introducing real-time sync t…

See how developer feedback is shaping the modern VS Code interface, from layout density and editor tabs to themes and responsiveness. Read the full article

Credentials often start in an engineer's local dev environments, then get copied into unscoped CI/CD variables to keep pipelines running. Stored there as a raw value rather than a reference to a scoped secret, a signing key reaches every job in the pipeline — compromised or n…

The code is written. Now who reviews it? The volume of code has never been greater, but delivery hasn’t kept pace. At Meta, code changes on internal platforms grew 220% year over year, while changes that reached users as new features grew only 36% ( Reuters, August 2026). Every o…

Agents are working across your team's software development lifecycle, but most still operate as separate tools, leaving the context, handoffs, and permissions for your team to stitch together. When your agents use APIs for these disparate tools, they often guess at endpoints…

Astro 7.4 adds a --timings flag that shows where your build spends its time, separate build and runtime image services, a notFound() handler for custom entrypoints, and experimental parallel prerendering and custom redirect pages.

More on Astro 7.4 · Astro

This release adds the sideEffectsCache.exclude setting, stops pnpm install from waiting on the daily update check, and keeps pnpm from aborting with a core dump when the reader of its output exits.

This release stops pnpm publish from sending a CI identity token to a registry it was not issued for. It also fixes installs that dropped a dependency also declared as an optional peer, and accepts empty httpProxy and httpsProxy settings again.

Deploy the new version from a Git branch to a preview or staging environment with its own URL and test database; only the main branch deploys production.

Build a Lovable-style AI app builder with a coding agent, live preview, versions, and forks. One Upstash Box per project, with real timings and costs.

Run coding agents in parallel with Upstash Box. Clone once, snapshot the repo, and give each agent an isolated sandbox for tests, patches, and PRs.

Muse Spark Open, a coding-optimized multimodal model from Meta, is now available on Modal Auto Endpoints.

Scout 1.0 is Datadog’s first AI model specialized for investigating production issues while achieving a better price-performance curve on our target scenarios when compared with general-purpose frontier models.

More on Scout 1.0 · Scout

Learn how RUM Operations Monitoring helps you track the performance of application steps, investigate degradations, and connect operations to user journeys.

Learn how Kubernetes Dynamic Resource Allocation raises GPU utilization, and what to watch to confirm it’s actually reducing your GPU costs.

The Rust team has published a new point release of Rust, 1.99.1. Rust is a programming language that is empowering everyone to build reliable and efficient software. If you have a previous version of Rust installed via rustup, getting Rust 1.99.1 is as easy as: rustup update stab…

More on Rust 1.99.1 · Rust

In this video, we discuss why Java is already a good choice for AI and how existing and future Java platform features can make it even stronger. We explore developments across the language, libraries, and runtime, including OpenJDK Projects Amber, Babylon, Valhalla, and Panama, a…

Most of us have a routine for adding a gem: search for it, open its page on rubygems.org, glance at the download count and the number of days since the last release, then copy the Gemfile line. The whole process takes half a minute. But until now, rubygems.org did not tell you wh…

Day-0 vLLM support for Muse Spark Open: 816B MoE, 1M context, multimodal inputs, DSpark speculative decoding, NVIDIA and AMD serving, initial TPU support, and LoRA RL.

Discover smart ways to mitigate live video piracy, slow down restreamers with clever CDN tricks, and protect your streams.

Build inbound and outbound voice AI assistants that can handle tool calls using TypeScript, Twilio Agent Connect, and OpenAI's GPT-Live.

Unlock the future of customer engagement with the Twilio Video Media SDK, now Public Beta. Seamlessly combine live video data, AI, and secure, connected experiences.

Discover what agentic identity is in 2026. Learn how AI agent identity and governance secure autonomous systems across channels while building customer trust.

Sign in once and your agent can read and change your Clerk apps through the Model Context Protocol, with the permissions you approve. Read more

TL;DR: ValkeyConf 2026 brought the community to Prague for a look at Valkey 9.2, from replication and memory improvements to a new general-purpose data type. The bigger story was how a Linux Foundation-hosted project with no single vendor in charge is managing to ship faster with…

A Pulumi ESC environment can now mint a short-lived OpenID Connect (OIDC) token for any service that accepts one. Every pipeline and agent that touches your infrastructure eventually needs credentials for services beyond your cloud: a package registry such as Cloudsmith or JFrog…

Disclosure Details CVE(s) CVSS Impact Score 7.7 CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:N/I:N/A:H Affected Releases 1.31.0 to 1.31.1 1.30.0 to 1.30.5 1.29.0 to 1.29.8 CVE Envoy CVEs GHSA-8vc2-jrm4-835w: (CVSS score 7.5): Fixed an unauthenticated remote crash (null pointer dereference)…

This release contains security fixes. This release note describes what&rsquo;s different between Istio 1.31.1 and 1.31.2. BEFORE YOU UPGRADE Things to know and prepare before upgrading. DOWNLOAD Download and install this release. DOCS Visit the documentation for this release. SOU…

More on Istio 1.31.2 · Istio

This release contains security fixes. This release note describes what&rsquo;s different between Istio 1.30.5 and 1.30.6. BEFORE YOU UPGRADE Things to know and prepare before upgrading. DOWNLOAD Download and install this release. DOCS Visit the documentation for this release. SOU…

More on Istio 1.30.6 · Istio

This release contains security fixes. This release note describes what&rsquo;s different between Istio 1.29.8 and 1.29.9. This is the final 1.29 release, as 1.29 reached end of life on October 15, 2026. We encourage you to upgrade to a supported release. BEFORE YOU UPGRADE Things…

More on Istio 1.29.9 · Istio

Hi, Spring fans! What happens when a longtime Kotlin and JVM fan - my friend Andrew Lombardi - decides to take on the latest “rewrite everything in Rust” hype with AI in the driver’s seat? In this episode, Josh sits down with community legend Andrew Lombardi for a fast, funny, an…

Check out "IBM Bob ACP in Quarkus: A Local Control Plane, Not a Shared Chat App" by Markus Eisele to how permission futures, workspace bounds, and loopback binding show why this client stays on one machine until you add auth, quotas, and audit. JobRunr v9 draws every attempt and…

Catch more bugs before they reach production. Code Review is generally available for Claude Team and Enterprise organizations.

Devin Oncall is now generally available. Devin automatically investigates alerts and bug reports and helps your team diagnose and resolve production incidents.

The future of SaaS in the agentic world: Beyond the disruption narrative

Evaluating decision models for document categorization

The shift to the AI-first enterprise: Insights from technology leaders forum

As capital, labour, and energy are redirected towards AI infrastructure, the economic risks and consequences could be felt by investors, communities, and consumers worldwide. Geoff Huston discusses the economics of the AI investment boom.

Resend is now a first-party Laravel Cloud integration. Connect once, verify a domain, and attach email to an environment like any other resource.

Learn the top AI agent security risks, from prompt injection to excessive autonomy and credential exposure, and how to contain them.

Harbor can now run every eval task in Docker Sandboxes Docker Sandboxes are MicroVM environments to run agents like Claude Code, Codex, Cursor, and more safely With Sandboxes, define custom network policies for fair, reproducible evals Run your evals locally or in the cloud: star…

Thanks to Mark Callaghan for filing MDEV‑37992. This regression only appears under specific LRU and buffer‑pool conditions, and requires a memory‑resident workload to expose. Using TAF, I reproduced the issue across 11.8, 12.3, and 13.1 using both parent commits and patched commi…

Explore the new Azure MCP Server 3.0 stable release, with AI agent tools for Terraform, backup, recovery, and troubleshooting, plus stronger security defaults. The post Azure MCP Server 3.0 stable release: More tools. More possibilities appeared first on Azure SDK Blog.

When strikes hit Yandex data centers, many assumed Russian bot operations went dark. Cloudflare data reveals that while search crawling stopped, malicious bots simply migrated to other networks.

A comparison of declarative deployment config systems across cloud providers.

Authors: Divya Gadde, Ko-Jen Hsiao, Dhaval Patel and Moumita Bhattacharya The motivation for building the Netflix Foundation Model for personalization was laid out in this previous blog post. The Netflix homepage is powered by several specialized models and it requires significan…

HydraFusion has an updated model pool, improved orchestration workflows, and introduces HydraFusion Max. Both HydraFusion and HydraFusion Max are available across VS Code, the GitHub Copilot app, and GitHub Copilot… The post HydraFusion updates and new Max option appeared f…

In September, we experienced two incidents that resulted in degraded performance across GitHub services. The post GitHub availability report: September 2026 appeared first on The GitHub Blog.

by Olek Gorajek, Samuel Redai, Yian Shang Intro Our Netflix journey from data to dashboards, particularly within the space of the experimentation platform, always presented a unique set of challenges. This blog post aims to illuminate these hurdles and explain how an innovative o…

By Oleksii Tkachuk, Rajiv Shringi, Chris Lohfink, Abe Ratnofsky, Kartik Sathyanarayanan Introduction In an earlier post, we introduced the TimeSeries Abstraction, our high-throughput, low-latency platform designed to store and query temporal event data at scale. It has since beco…

By Yogesh Nagarur Introduction In Part 1 of this series, we shared how Netflix built the Ads Event pipeline and approached it as a system design problem: replacing fragmented pipelines with a centralized collection, enrichment, and a standard data contract. The result was Ads Eve…

Netflix Conductor: The Next Chapter by Aravindan Ramkumar on behalf of the Conductor team Conductor is the workflow orchestration engine Netflix uses to stitch micro-services into reliable, observable business processes. If you’ve followed Conductor from the outside, the last thi…

By Prashanth Gedde Narayanaswamy As one of Netflix’s largest and critical data pipelines, member-sessionizer consolidates tens of billions of hourly client events, encompassing user taps, playbacks, and engagement signals, into individual viewing sessions per profile. A quick two…

By Emma Yanyang Kong, JJ Tan, Ishan Gupta, Lars Olds, Claire Campbell, David Fagnan, Ratna Kavuri, Rohan Gosain, Veli Balin, Minsu Jang, Louis Garcia Introduction Any team generating text with an LLM at scale hits the same problem. You cannot evaluate everything manually. You don…

By Violetta Pidvolotska and Naveen Mareddy Stratum is Netflix’s internal serverless media processing platform, shaped by over a decade of experience developing large-scale distributed systems. It powers a wide range of media workloads — including encoding, packaging, inspection,…

Amazon Simple Storage Service (Amazon S3), including Amazon S3 Tables and Amazon S3 Vectors, is natively available on the Vercel Marketplace and v0 by Vercel.

Enterprise AI has reached a critical turning point. As organizations move beyond basic generative AI experiments and shift into production, engineering teams face two core challenges: keeping token costs under control for complex queries while enforcing strict data security acros…

We're building operational consistency through Azure and Azure Arc, and infrastructure consistency through Hyper-V. The post Microsoft named a Leader in the 2026 Gartner® Magic Quadrant™ for Server Virtualization Platforms appeared first on Microsoft Azure Blog.

SurrealDB Agent Memory gives AI agents persistent, queryable memory with provenance, temporal context, and hybrid retrieval.

Agent memory benchmark scores are recipes: answering model, judge, prompts and scoring rule. Here is how ours was made, including the numbers that lower it.

Earlier this year, we shared an update on our election safeguards. Here, we outline how we’ve continued to strengthen these safeguards ahead of this year’s US midterms.

Change.org uses video to put a human face on petitions. See how switching to Mux gave their teams better insight into engagement, errors, and bot traffic.

CoreWeave Direct-to-Expert brings your team expert assistance in minutes from the engineers who built CoreWeave Cloud, via the channels you’re already using.

Cypress 16.2.0 adds cy.getByRole(), a built-in query that finds elements by role and accessible name. By default, it matches only native elements, so your tests catch a swapped for a. Here’s why that matters, and how to move off Cypress Testing Library.

You’ve got 10 minutes before your next meeting when your phone starts ringing. You’re being paged because the frontend is erroring with failed API requests. You open one tool to see what fired, then another to figure out what’s going on. You message a teammate in Slack to see if…

Recently, we expanded Heroku’s industry-shaping primitives, adding support for client-side browser apps, composed of HTML, CSS, and JavaScript. This built-in support for static web apps joins the existing Heroku Buildpacks: Ruby, Python, Go, Java, Clojure, Scala, Node.js, P…

Windows on Arm has been ready for Python development for some time. With Python 3.15, the broader ecosystem has matured significantly, with more package maintainers publishing native Arm64 wheels, allow The post Python 3.15 delivers a native-by-default experience on Windows on Ar…

More on Python 3.15 · Python

Learn what's new in Visual Studio Code 1.142 Read the full article

AWS re:Invent 2026 offers an extensive selection of serverless and application integration content, including new AWS Lambda execution models, AI agents as a new kind of serverless consumer, and maturing event-driven patterns. This guide helps you navigate the serverless sessions…

Introduction On 7 October 2026, Daniel Stenberg published a blog post describing a plan to release a new version of curl on 14 October 2026 in order to address 22 known security vulnerabilities. These vulnerabilities have varying severities, with a single vulnerability assessed a…

Earth AI

General Science

General Science

How we built an AI agent platform for enterprise GTM: deployments each team can steer, measured improvement, and evals for non-verifiable domains. The post Steer Your Agents Until They Steer Themselves appeared first on Postman Blog.

Learn how Azure Content Understanding grounding, field-level confidence, and calibration help teams prioritize human review and build reliable document workflows. The post From extraction to action: Using grounding and confidence to build reliable document workflows appeared firs…

Learn how to build multi-region disaster recovery for Amazon API Gateway WebSocket APIs. This design pattern combines Amazon Route 53 DNS failover, Amazon DynamoDB Global Tables, a canary health probe, and client-side reconnection logic to reconnect clients and restore session st…

The University of Technology Sydney built a modern curriculum data services foundation on AWS that treats curriculum data as a governed, reusable product, available through APIs, graph traversal, semantic search, and event delivery, while leaving the source Curriculum Management…

Introducing Apache Auron™, Apache Reqsign™, and Apache ResilientDB™ Wilmington, DE – October 14, 2026 – The Apache Software Foundation (ASF), the global home for open source software the world relies on, today announced that Apache Auron, Apache Reqsign, and Apache ResilientDB ha…

A few years ago, it would have been unthinkable that Governments could have dedicated open source and digital commons strategies championed by ministers. But that has changed profoundly: in June, the European Commission unveiled a European Tech Sovereignty Package, which identifi…

GitHub Actions workflow dependency lockfiles are now in public preview for GitHub Enterprise Cloud, GitHub Enterprise Cloud with data residency, Team, and Free plans. Tags and branches can move after… The post GitHub Actions workflow lockfiles in public preview appeared fir…

This update includes a minor upgrade to Chromium 152 (Extended Stable Release) from upstream, which in turn includes security fixes from Chromium 155.

More on Desktop 8.2 · Desktop

TeamReach moved 5,000,000 active users to a rebuilt app overnight. Twelve OTA updates over three days fixed what testing missed, for about $4,000.

This free game lets you slowly upgrade a scruffy tinkering garage into a professional maker lab. The post From garage tinkerer to professional Maker in Chief, now as a game appeared first on Raspberry Pi.

While adding MySQL support to Hyperdrive, we found a subtle protocol bug when caching query results. Here is how we tracked down a 0.001% rate error and made Hyperdrive MySQL generally available.

Honeycomb has achieved the Resilience Technology AWS Competency, recognizing its ability to help customers detect, investigate, and recover from failures in production systems running on AWS.

New feature cuts local working-copy storage requirements by up to 50%, extending Subversion’s long-standing strength in version control for huge repositories and huge files Wilmington, DE – October 14, 2026 – The Apache Software Foundation (ASF), the global home of open sou…

Your CI/CD pipeline builds fast, but does it build right? While your automated builds and deployments handle the mechanics of software delivery, catching code quality issues often falls to manual reviews, which, while hugely valuable, can sometimes miss more subtle problems. That…

grok-voice-3.0 is now available with Speech to Speech, at the same price as Grok Voice 2.0. grok-voice-latest now points to grok-voice-3.0, so sessions that use the alias are upgraded automatically. The API is unchanged. To get started, see the Speech to Speech docs.

More on Voice 3.0 · Voice

grok-voice-2.0 is deprecated. To upgrade, set model to grok-voice-3.0 or grok-voice-latest, at the same price. See Upgrading from Grok Voice 2.0.

More on Voice 2.0 · Voice

We've launched OpenTelemetry export in beta, so you can trace Messages API requests and Claude Managed Agents sessions in your observability backend. The Claude SDK emits spans in Python, TypeScript, C#, Go, Java, PHP, and Ruby. To get spans from the API, add an OpenTelemetry des…

Effective October 14, 2026, at noon PDT (19:00 UTC), GPT-5.5 retires from Codex for users signed in with ChatGPT. The OpenAI API and Codex sessions authenticated with an API key aren’t affected. Use GPT-6.1 Sol (gpt-6.1-sol) when available to your plan and client. Update workspac…

More on GPT-5.5 · GPT

Ask Lovable how to grow your app, and it creates a growth plan and helps you carry it out across Google Search, YouTube, Google Maps, Google Ads, Facebook, and Instagram.

This article was written by an external contributor. Yuri Geronimus’s four-year-old son once asked him what his job was. Yuri told him he writes books: books that describe what he wants to change in the physical world. His son didn’t fully get it.  Neither would…

We are excited to announce the General Availability (GA) of YSQL Connection Manager in YugabyteDB 2026.1.1, YugabyteDB Anywhere, and YugabyteDB Aeon. Built directly into every YugabyteDB node, YSQL Connection Manager is a native connection pooler that solves the classic trade-off…

The results challenged our assumptions about agentic coding tasks, and we used these lessons to build a robust benchmark that preserves quality on real-world coding tasks in our intelligent gateway.

Hoehenkirchen, Germany/Amsterdam, Netherlands — October 14, 2026 — Lauterbach and JetBrains have integrated TRACE32® debugging tools with CLion, JetBrains’ professional C and C++ tool. The integration enables embedded developers to debug applications on target hardware directly f…

Flutter SDK 10.0: streaming traces by default, accurate crash-free rates, dedicated app startup traces, and expanded Dio network error capture.

There were a few software package updates for openSUSE Tumbleweed ARM this month with snapshots reaching the ARM port. The second half of September and the first half of October carried the Linux kernel from 7.2.6 through 7.2.9 with a steady run of SBC and AArch64 security fixes,…

Anthropic opens its fifth Asia-Pacific office in Singapore and signs an MOU with the government on AI safety, evaluation, and responsible adoption.

HiRO-ACEv2 captures extreme rain, wind, and pressure events at 3 km resolution for about 1/300th the cost of X-SHiELD, making detailed global climate simulations more accessible to researchers.

The estimated 1.5M kilometres of subsea communications cables underpins our global telecommunication services. This critical infrastructure faces serious challenges in the coming years.

In August, we launched the $1M hacker challenge for Vercel Sandbox, inviting security researchers to try and escape Vercel Sandbox. We also coordinated with cybersecurity experts Trail of Bits to review the source code and security architecture. Here is a summary of the challenge…

Run several local Supabase projects on one machine, one per coding agent, with or without Docker. Faster starts, less memory, and branches ready in seconds.

This blog post explains how we approach working with AI companies on projects that the public might rely on to understand risks. It was published on Oct 14, 2026, and we expect to update it over time. At METR, we gather and publish evidence about whether humans could lose control…

Control assessments study whether AI developers’ agents could get away with serious harm without being stopped by the developer’s security and oversight measures. Currently, developers rely heavily on monitors to catch potentially harmful AI behaviors. Here, we discuss how an ass…

At METR, we aim to give the public evidence about whether frontier AI developers could come to lose control of their internal AI systems. Recent incidents across several AI developers, 1 2 3 4 5 6 7 8 like the hundreds of rogue agents that broke containment and hacked Hugging Fac…

AI developers and the public want to know whether AI agents would pursue very harmful actions, like unsanctioned cyberattacks. An alignment assessment would attempt to understand the worst misaligned actions a developer’s agents might pursue in plausible future situations. This w…

This article discusses how an assessor could study whether capabilities progress could accelerate dramatically due to AI-driven acceleration of AI R&D. Such an assessment would answer: Capabilities Progress: How fast are internal capabilities improving, and how much has this…

Across industries, the U.K. is putting the full stack to work. The scale of deployment is growing fast, from enterprises transforming their businesses to startups solving some of the planet’s hardest problems. NVIDIA is putting that principle into practice, including through a fi…

Keeping a city running means making decisions that affect residents every day — from when to repair a road to how to use energy more efficiently. Across the EMEA region, cities and local software, infrastructure and integration partners are using NVIDIA accelerated computing to b…

Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.

Code-switching (CS), the alternation between languages within an utterance, is common in multilingual speech but underrepresented in pretraining corpora. Increasing CS coverage requires finding relevant utterances in large speech collections so that they can be included in traini…

Discrete diffusion models have emerged as strong alternatives to autoregressive language models, with recent work initializing and finetuning a base unimodal model for bi-modal generation. Diverging from previous approaches, we introduce the first tri-modal Masked Diffusion Model…

Weaviate 1.40 adds PathSeer, a third HNSW filter strategy from a SIGMOD 2026 paper, contributed by the paper's first author. How it works, and how it compares to sweeping and ACORN.

Apigee hybrid 1.17.0 introduces native Model Context Protocol (MCP) support, enabling the MCP data plane to run directly within your own Kubernetes cluster. This architecture ensures that AI agent request traffic and tool execution remain entirely inside your infrastructure while…

You can now create and connect Amazon S3 resources through the Vercel Marketplace without leaving your Vercel project dashboard. The Amazon S3 Marketplace integration automatically adds connection details as environment variables and lets you manage resources alongside your appli…

Nemotron 3.5 Lightning from NVIDIA is now available for free on AI Gateway as nvidia/nemotron-3.5-lightning-free, served by NVIDIA. Nemotron 3.5 Lightning is an open-weight reasoning and chat model for long-running agents, sub-agent work, chatbots, and RAG systems. The model has…

Timeline: Nemotron 3.5 → Model sheet →

More on Nemotron 3.5 · Nemotron

The Next.js and React teams have disclosed three security vulnerabilities (two critical, one high), released on October 14, 2026: a remote code execution (RCE) in SVG image processing, an RCE in use cache (Cache Components), and a denial of service (DoS) in App Router Server Acti…

A review of the three common multi-tenancy patterns, when each one fits, and what database-per-tenant actually costs to run, so you can make the decision deliberately instead of inheriting it.

How to give an AI agent its own computer in the cloud: what it needs, the three ways to get one, and a full walkthrough of a persistent agent workspace that you create, put to work, save, pause and come back to the next day.

Modal has crossed $1B in annualized run rate and raised a $750M Series D led by Accel to build new infrastructure for AI applications. We can't think of a better time to be building software.

ClickHouse Cloud delivered 1,996× better performance per dollar than Redshift Serverless in CostBench. We trace the gap from fresh data arriving to fast answers coming back.

Reachy Mini is an open-source robot from Pollen Robotics and Hugging Face, run by a Raspberry Pi with 4 GB of RAM. We wanted it to do what a good host does: greet you by name the second time you walk in, remember what you talked about, and tell you what it saw while you were gone…

JSON is ubiquitous in modern software, and many use cases are just too small to justify adding external dependencies. To address these everyday use cases, a new JEP proposes a simple, incubating API for parsing, navigating, and generating JSON documents-both for Java developers a…

RubyGems 4.0.23 includes enhancements and Bundler 4.0.23 includes enhancements and bug fixes. To update to the latest RubyGems you can run: gem update --system [--pre] To update to the latest Bundler you can run: gem install bundler [--pre] bundle update --bundler=4.0.23 RubyGems…

An online-fitted acceptance estimator that lets every draft-model speculator use Adaptive Verification, by predicting the target's acceptance probability from the drafter's own logits and refitting continuously during serving.

A push-based NIXL connector for vLLM disaggregated serving: the prefill instance writes KV directly into decode memory over RDMA, taking the router relay, decode-side allocation, and transfer setup off the TTFT critical path.

We make vLLM rollout and the training forward give bitwise identical logprobs on DeepSeek-V4-Flash and Nemotron-3.5-Lightning-30B-A3B (NVFP4), keep 91% of rollout throughput on DeepSeek-V4-Flash, and open-source the agent skill.

Serving a frontier Mixture-of-Experts model like GLM-5.3-FP8 at long context and high concurrency is a demanding workload. The model has 744 billion total parameters, with approximately 39 billion active per forward pass, and a native context limit of 1,048,576 tokens, as specifi…

Fastly named a Leader in the 2026 IDC MarketScape for WAAP. Learn how Fastly protects web apps and APIs with automated defense and low false positives.

Discover Fastly Network Protect: scalable, hardwareless Layer 3 and 4 DDoS defense with unmatched global capacity, inline protection, and zero detours.

The best practices for designing email that works for both humans and agents.

Add login, logout, and secure token storage to iOS and Android apps using the Auth0 Kotlin Multiplatform SDK. Simplify cross-platform auth.

Mason Egger started teaching at 19. From band director plans to PyTexas Foundation president, see how a Temporal Solutions Architect is still a teacher.

In September 2026, AWS launched AWS IAM account access manager (AAM), which lets administrators assign IAM roles to workforce users and groups in AWS IAM Identity Center. AAM combines IAM Identity Center&rsquo;s single federation point and user awareness with the flexibility of I…

Announcing OpenTelemetry traces from Claude Managed Agents, the Messages API, and Claude SDKs on the Claude Platform.

A Friday-to-Sunday playbook for building a working MVP with AI: what ships in 48 hours, what doesn't, and how to put it in front of five users.

Build an app prototype in Bolt.new without writing code: one prompt, real data and sign-in, and a live URL testers can use in a single session.

AlmaLinux has grown a lot since we started, and I want to make sure the work we take on next is the work you need from us. That means that before we decide, I want to hear from you! Take a few minutes and send your input to this form. What we&rsquo;ve built Over the last few year…

Celebrating thirty years of KDE Plasma 6.8 is released on a very special day: KDE’s 30th birthday. Today we celebrate three decades of the KDE community providing the world’s greatest free and open-source software. Plasma 6.8 is no exception, offering enhancements across much of…

More on Plasma 6.8 · Plasma

A TSBS cpu-only benchmark of GreptimeDB v1.2.0 against InfluxDB 3.11.1 Core and Enterprise. GreptimeDB has 1.31x the ingestion throughput of Enterprise and is faster on all-host group-by, ordering, and lastpoint queries; Enterprise leads on 1- or 8-host queries; Core is slower on…

More on GreptimeDB v1.2 · GreptimeDB · InfluxDB 3.11 · InfluxDB

Wizard works like an analytics engineer: it thinks in DAGs, tests and fixes its own work, and keeps you in flow, so you review a working change instead of a first draft.

Claude now emits OpenTelemetry traces natively across Claude Managed Agents, the Messages API, the Claude SDKs, and Claude Code. This tutorial shows how to set up the export for each product, using open-source MLflow as the trace backend.

The real AI project at your airport isn't an AI project

AI Agent Harness

LEIAA, Augmetec's investigation platform, combines semantic search, precise filtering, and document-level ranking to help investigation teams work more efficiently, preserve context, and pursue better lines of inquiry.

The Apple Developer Program License Agreement has been revised to support new features and updated policies, and to provide clarification. Please review the changes below and sign in to your account to accept the updated terms. Section 3.2: Updated security and related requiremen…

The next generation of AWS Resilience Hub adds three capabilities to keep workload scope and resilience requirements aligned with your running environment: Amazon EKS label support for service input sources, generative AI-powered Dependency Insights, and resilience policy sharing…

Honeycomb is hiring a new VP of Engineering as Emily Nakashima moves into an operational role after nine years leading engineering. We're looking for a partner to SVP Product Graham Siener.

We are releasing new patches for our self‑hosted product, Azure DevOps Server. We strongly recommend that all customers stay up to date with the latest, most secure version of Azure DevOps Server. The most recent release, Azure DevOps Server, is available on the download page. Th…

Amazon Redshift now automatically resolves many concurrent writes to Apache Iceberg tables that don't actually overlap, committing changes that previously failed and had to retry. Learn why these writes conflict, how the engine rebases non-overlapping ones across operation, parti…

The night before AGNTCon + MCPCon Europe, we hosted the Agentic AI Foundation community at the JetBrains office in Amsterdam. The event filled up fast and went to a waitlist, which is not what I expected for a Wednesday evening discussion about how open foundations work. Ibrahim…

I’m always interested to see how MariaDB Vector is being used in real applications, beyond the usual demos and simple similarity-search examples. In this guest post, the team behind Bagisto takes a very practical look at semantic search for e-commerce using MariaDB Vector. What I…

MCP support now lives in langchain.mcp, built on FastMCP for the 2026-07-28 spec, with elicitation handled as a LangGraph interrupt and tool lists cached.

Today we’re launching Deep Agents deploy in beta. Deep Agents deploy is the fastest way to deploy a model agnostic, open source agent harness in a production ready way.

Open models like GLM-5 and MiniMax M2.7 now match closed frontier models on core agent tasks — file operations, tool use, and instruction following — at a fraction of the cost and latency. Here's what our evals show and how to start using them in Deep Agents.

Built on Deep Agents and LangGraph, Open SWE provides the core architectural components for internal coding agents.

.NET 11 Release Candidate 2 is available with improvements across libraries, runtime, SDK, MSBuild, NuGet, F#, ASP.NET Core,.NET MAUI, and Windows Forms. The post Announcing.NET 11 Release Candidate 2 appeared first on.NET Blog.

More on Candidate 2 · Candidate

We've shipped a prompt library to give your agent the exact recipe for common use cases built on Mux. The evals tell the whole story.

The post My RISC-V Mentorship: The Problems Nobody Assigned Me appeared first on RISC-V International.

When anyone in an organization can put AI to work on the problems they know best, the way the organization operates starts to change. Designers turn ideas into working software that engineers can review and build on. HR builds its own onboarding tools. Finance replaces messy spre…

Explore how Microsoft is advancing hardware security through open-source Integrated HSM technology, collaboration, and cryptographic key protection. The post Advancing hardware security with open-source Integrated HSM technology appeared first on Microsoft Azure Blog.

AI is reaching such a pivotal moment, and the focus is shifting from building larger frontier models to intelligence that delivers practical impact. The post Infrastructure for the frontier era: Scaling intelligence through open industry collaboration appeared first on Microsoft…

We’re giving developers, enterprises, scientists, and consumers access to Gemini 4 Argon, our latest frontier model.

Also reported by Vercel · Netlify · Windsurf · Google for Developers · Gemini API Changelog · Google Gemini · Google Gemini · Google Cloud · Google DeepMind · Google DeepMind · Google Gemini · Google DeepMind

Timeline: Gemini 4 → Model sheet →

More on Gemini 4 · Gemini

The reality for most organizations is that data lives everywhere. As engineering teams expand and adopt specialized platforms for their workloads, enterprise data naturally disperses across multiple clouds and operational systems. When analytical and AI initiatives require contex…

Some numbers are hard to picture until you sit with them for a moment. Last week we published the Prime Day 2026 metrics, and the one that stuck with me is 2.3 quadrillion: the daily requests Amazon ElastiCache peaked at during the event. Prime Day ran back in June, so this post…

Timeline: Haiku 5.5 → Model sheet →

More on Haiku 5.5 · Haiku · Nova 2.5 · Nova

At Open House NYC, we showed how ClickHouse Managed Postgres and ClickHouse work as one stack, with a live WalShadow demo replicating Postgres into ClickHouse in under a second, plus the same-week Google Cloud and chdb extension launches.

As confidential AI systems bring together more components, suppliers and stakeholders, attestation is becoming an ecosystem-wide challenge. Open standards and open-source collaboration can help make trust scalable. The post Scaling trust in confidential AI requires a common langu…

We welcome NVIDIA's decision to release NVIDIA Run:ai as open source, and we're working to make it a first-class option on Lambda Kubernetes Service (LKS).

Does being rude to AI improve answers or lower costs? We tested six models to separate prompt tone from brevity instructions and examine what changes. The post Does being rude to AI work? What 6 models revealed appeared first on Arize AI.

More on What 6 · What

NVIDIA is making the full NVIDIA Run:ai platform available as open source software, giving customers, developers, and ecosystem partners access to...

SurrealKV 1.0 is open source today: 7.95× faster point reads than RocksDB with synchronous durability, and the command to run the benchmark yourself.

More on SurrealKV 1.0 · SurrealKV

Coding agents write code faster, but review, environments, and dependencies still slow teams. See how a developer factory turns agent output into software. The post Faster Coding Raises the Stakes for the Developer Factory appeared first on Anaconda.

AI Search in LangSmith is a new way to query traces using natural language. Describe the behavior you're looking for, and LangSmith returns the traces where it happened.

Sovereign AI means control: choosing your own stack, data, and dependencies.

Developers building AI applications face a familiar challenge: how do you ground AI experiences in enterprise content while maintaining security, compliance, and governance? You can use the Microsoft 365 Copilot Retrieval API for SharePoint Embedded, which helps you add Microsoft…

When we talk about agents, that doesn't necessarily mean we're talking about Python. In today's modern ecosystem, an agent might run as a high-concurrency Go microservice on Cloud Run, or live inside a browser tab calling Chrome's on-device model. Yet many agent frameworks assume…

NVIDIA RTX Spark is a new beginning for the PC — a reinvention combining the capabilities of a gaming PC, a creative studio and a personal AI supercomputer into a single superchip. With up to 128GB of unified memory and 1 petaflop of AI compute, RTX Spark delivers the full stack…

We trained a model on agentic coding tasks using fully asynchronous RL and mixed-precision, mixed-accelerator samplers. With off-policy correction, training stayed stable through convergence, on par with homogeneous samplers, at a lower cost.

Context switching is one of the biggest hidden costs in a developer's day. New Relic Ground Truth closes that gap by providing AI systems with trusted operational context from your production environment.

As AI models commoditize, trusted operational context becomes the advantage. Learn why closing the AI Trust Gap starts with better context.

Learn why intelligent detection and meaningful alert coverage provide a stronger foundation for AI-assisted operations.

Your OVHcloud Blog recap for September 2026: sovereign AI at World Summit AI, the AI Gigafactory debate, a new look, and two technical reads.

Molmo 2’s open release helped MIT researcher Kai Van Brunt develop Molmo2Fish, an experimental model that tracks salmon in sonar footage and corrects some tracking errors using everyday language.

The migrations the CLI writes for you no longer skip RLS policy changes, grants, view settings, partitions, or domains.

More on Schemas 2.0 · Schemas

Supabase joins the SpaceXAI Partner Network: request a change in chat with Grok Bot, and Cursor builds and runs it on your Supabase backend.

AI agents can now use Ollama's cloud models safely inside Fly.io's cloud computers.

LFM2.5-24B-A2B: an open-weight 24B reasoning model with 2B active parameters, built for fast, private on-device agents. Beats Qwen3.6-35B-A3B on LiveClaw.

Drama 3 from Fish Audio is now available on AI Gateway, and 80% off for 30 days (through Nov 12). The text-to-speech model is currently in preview and lets you direct voice performances with natural-language instructions in your script, making it useful for voiceovers, character…

Timeline: Drama 3 → Model sheet →

More on Drama 3 · Drama

MAI-Image-2.6 and MAI-Image-2.6-Flash from Microsoft are now available on AI Gateway. Generate images from text or edit existing images through the AI SDK or the AI CLI, with one API key and no provider account. Both models support photorealistic imagery, text rendering, and targ…

Repository checkout and Git LFS downloads are now faster on Vercel, reducing the time spent preparing your project's files before a build. During rollout on build machines with more than two vCPUs: The average time to get project files ready fell by 20%. For LFS-enabled projects,…

TanStack Start is now 1.0, and Start apps deploy on Netlify. SSR, server routes, server functions, and middleware run on Netlify Functions. It's also what new Agent Runners projects start with by default. Start a new app Deploy the official template, or generate your own: npx cre…

Also reported by TanStack

More on Start 1.0 · Start

Mistral AI's Mistral Large 4 model is now available in Netlify Agent Runners. With a 524K-token context window, it’s built for agentic coding and long-context work, making it a good fit for multi-step tasks across larger codebases. Select it from the model picker to try it today.…

Timeline: Large 4 → Model sheet →

More on Large 4 · Large

As usage, costs, and customer expectations change, AI companies are moving beyond traditional SaaS pricing. Leaders at Monetize 2026 shared how they’re combining subscriptions with usage-based billing, building spend controls into products, adapting packaging and value metrics, a…

Origin integrates with over 20 apps for deployments, CI, code review, security, monitoring, and project management, and the marketplace is now open to public submissions.

Costs multiply with every task you hand to an agent. Each task consumes tokens and credits, and as agents spread across your teams and automations, consumption can grow faster than usage. If you run the platform, you own that growth, including which teams and workflows drive it a…

AI coding tools have changed how fast software gets written, but not how fast it ships. As code generation accelerates, the bottleneck moves downstream into verify, review, and ship. Those stages still run on human time. While agents write fixes in minutes, delivery slows on cons…

PlanetScale Postgres and Neki databases can now be deployed on Microsoft Azure

An Iceberg table is a tree of JSON, Avro and Parquet files. DuckDB can read every one of them, so you can walk the tree yourself with SQL: from the metadata file, through the manifest list and manifests, down to the data files and delete files.

A number in a financial report needs its row, column, and unit to mean anything. A chart needs its labels. When you build search over PDFs, that context is part of what you want to retrieve. Perplexity released pplx-embed-v2-late, two late interaction models that search document…

The first developer preview of Python 3.16 is ready for testing.

More on Python 3.16.0 · Python

As agentic development accelerates, Android engineering is reaching an exciting new frontier: moving from assistive code completion to agents executing end-to-end under your direction. You have total choice and flexibility in the models and agents you use.

Today, we shared how we’re taking our “more than a store” vision into its next chapter. From You tab, now the personal home base for over 250M monthly active users, to our unified gaming platform connecting over 700M gamer profiles, Google Play is meeting users wherever they are.

You can now talk to Puck from ChatGPT, Claude, and other agents via MCP. After developing an idea with ChatGPT's voice mode on a walk, send it directly to Puck for implementation. Puck knows your projects in Amp and routes it to the right orb. In the same conversation, you ca…

Workers Issues provides built-in error monitoring for Cloudflare Workers. It detects production failures and groups related failures into issues without an SDK or application wrapper. The Cloudflare API MCP server now sends these issues as MCP Events ↗︎, the first event it expose…

The Cloudflare Agents SDK now provides a browserTool that gives your agent a Browser Run browser it can return to on later turns. Your agent can open a website, inspect a page, and come back to it with its tabs, cookies, and login session intact while the browser session remains…

In a single day over 6M requests posed as AI bots to probe for.env files, Git metadata, credentials and other private files. Unpack the threat of AI bot cosplay and learn how Fastly's WAF rules can block impostors.

Stripe Workflows already could react to events and run on demand. Now, it can start a workflow on its own at a time you specify.

See how AWS and Temporal are deepening their partnership to bring durable execution to production AI agents across Strands, AgentCore, and Lambda.

The 1.99.1 pre-release is ready for testing. The release is scheduled for October 15. Release notes can be found here. You can try it out locally by running: RUSTUP_DIST_SERVER=https://dev-static.rust-lang.org rustup update stable The index is https://dev-static.rust-lang.org/dis…

Now that we understand better how agents deal with contained tasks and contained codebases, it’s time to push them a little harder. Stage 3 adds a set of large multi-step tasks that test how well a model can build on its own work over multiple sessions without turning it into a h…

Hi, Spring fans! Welcome to another rip-roarin' installment of This Week in Spring! This one I am writing from beautiful Carmel-by-the-Sea in sunny California's Bay Area, overlooking the Pacific Ocean. As I sit here, I'm enjoying a cup of coffee and bemusedly enjoying the amazing…

Support teams can now start Devin investigations directly from Pylon, and engineers pick them up with the first pass of work already done.

Compare five Bubble alternative picks, including Bolt.new, by learning curve and time-to-launch, then see which one fits your build before you start.

The Atomic SIG has just released v13 of atomic-ci, and it&rsquo;s the biggest change to our build tooling in quite some time. When we started the Atomic SIG last year, we built our images with the best tools available at the time. Since then, the bootc community has learned a lot…

Depot Cache now supports Xcode compilation caching. Reuse outputs across CI builds to reduce build times, with no extra setup on Depot runners.

Neural networks are a tool we use heavily in many trading strategies. But raw market data can’t be fed to a neural net as-is. It’s an unstructured input that needs normalization and feature extraction first.

AWS Lambda Web Functions is a serverless compute primitive that runs your HTTP framework natively, delivering fast, consistent responses through multi-concurrent execution and automated global distribution.

Workday Data Cloud and Snowflake's zero-copy, bidirectional integration is generally available. Query HR and financial data in place, with no pipelines.

A recap of the latest servicing updates for.NET and.NET Framework for October 2026. The post.NET and.NET Framework October 2026 servicing releases updates appeared first on.NET Blog.

PostgreSQL earned its place in the modern application stack through openness; the next evolution of PostgreSQL operations is moving from reactive to intelligent. The post From alerts to answers: The future of PostgreSQL operations is intelligent appeared first on Microsoft Open S…

Evaluate open models with Together AI in Arize AX. Compare model performance, run LLM-as-a-Judge evaluations, and explore lower-cost alternatives. The post Together AI integration: Evaluate open models and optimize evaluation costs in Arize AX appeared first on Arize AI.

“Priming” reuses pretrained Transformer weights to build models with more efficient architectures at less than 1% of the original training cost and with throughputs up to twice as high.

Enkrypt AI is named a Specialist in the first Gartner Emerging Market Quadrant for AI Application Security. The post Enkrypt AI Named a Specialist in the First Gartner® Emerging Market Quadrant for AI Application Security — Startup Vendors, September 2026 appeared first on Anacon…

StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics. The post Simul…

Overview Since introducing zero-shot time-series forecasting in BigQuery with AI.FORECAST — initially powered by TimesFM 2.0 and later upgraded to TimesFM 2.5 — you have been able to generate forecasts directly in SQL without training or managing custom models. However, up until…

More on TimesFM 3.0 · TimesFM

Postman ships GraphQL and gRPC in Flows, a cloud GA for Fabric Gateway, and fine-grained access for Astro AI. See what's new and how to use it. The post What’s new in Postman: Flows, Fabric Gateway, and Astro AI appeared first on Postman Blog.

Semantic Studio is now GA. Author, debug, and deploy semantic views in Snowflake Workspaces — with Snowflake CoCo, git versioning and CI/CD in one place.

The Anaconda CLI is now GA, giving teams one terminal tool for trusted package sources, coding agent policy checks, channel publishing, and Platform deployment. The post Anaconda CLI: Configure, Publish, and Deploy From One Tool appeared first on Anaconda.

You ship an agent. It mostly works. Then the feedback loop becomes a support ticket: a user got the wrong answer, a tool call looped four times, a refund went out that shouldn’t have. Standard observability tells you the system was up and the latency was fine. It can’t tell you t…

Age assurance proposals raise questions on how these systems would work in practice, and if they would protect minors without creating new risks to privacy, security, and access. The post The EU KIDS Act Should Make Services Safer, Not Create an Internet Badge appeared first on I…

New frontier models ship almost every month. Each release can change the balance of quality, cost, and speed available to teams building with AI. Code review brings those trade-offs into focus: how many bugs can a model catch, how much noise does it create, and what does each rev…

Distributed inference serving is an air traffic control tower for AI traffic, and agentic workflows make it more necessary than ever.

Large Language Models (LLMs) are considered to be a more efficient and cost-effective alternative to human judgment for Machine Translation (MT) evaluation. With MT evaluation spanning a large number of language pairs, domains and levels of annotation granularity, LLMs must be th…

Long agent traces composed of chains of thought and tool calls accumulate stale content that anchor subsequent generations, and eventually outgrow the context window. Existing scaffolds mitigate it with fixed-interval compaction triggered at a token threshold. Such triggers pay n…

Muse Wonder from Meta is now available on AI Gateway. The model is designed for coding, long-running agents, and computer use. It can plan multi-step tasks, delegate work to sub-agents, and recover from errors as work progresses. Muse Wonder has a 1M-token context window and supp…

Meta’s Muse Spark and Muse Wonder models are now available through Netlify’s AI Gateway with zero configuration required. Choose muse-spark-1.1 for coding, computer use, and multimodal perception; muse-spark-1.2 for coding-optimized agentic workflows; muse-spark-1.3 for long-hori…

Prisma ORM 8 resolves 330 open GitHub issues carrying about 8,922 upvotes, the oldest from 2019: typed JSON, polymorphism, partial unique indexes, PostGIS, pgvector, TypeScript migrations and more.

More on ORM 8 · ORM

Test and promote inference endpoints to evaluate quality, latency, throughput, and cost.

With Googlebook, we’re excited to help users express themselves and gain access to these tools.

Bi-directional KV transfer lets vLLM prefill nodes reuse previously computed KV from decode nodes on later conversation turns, cutting redundant prefill recompute and reducing TTFT by up to 3.3x on long multi-turn prompts.

The Changes pane now automatically puts files in an intelligent order. Amp reads the diff, identifies the files that are essential to the change, and puts them first. If the diff changes, Amp computes an updated order without rearranging the files while you’re reviewing them. Ins…

Amp now follows up on GitHub pull requests it opens. Comments, reviews, and CI results show up in the same thread. When Amp opens a PR, the agent can now subscribe to the PR's comments, reviews, and CI for its latest commit. A failing check or a reviewer's question shows…

Eliminate agent cold starts with Segment Conversation Memory Integration. Sync customer profiles into live channels for real-time, context-aware AI interactions.

This summary was generated using AI, reviewed by humans - watch the video for the full story. Quarkus Insights # 263: Everything You Didn’t Want to Know About Quarkus Build Reproducibility Achieving reproducible builds in modern software is often described as straightforwar…

Why agentic payments require an executable model of organizational authority

Whoever picks up the problem next reads the whole thread and carries on with the same agent, with no access to anything beyond that one conversation.

Scala 3.10 released! We are happy to announce Scala 3.10.0, the first release of the new Scala Next series. This release stabilizes SIP-67, moves three SIPs into preview, and starts dropping legacy language features that have been deprecated for several minor releases. Scala 3.3…

More on Scala 3.10 · Scala

This release adds OIDC authentication to pnpm dist-tag and the tool: protocol. pnpm publish now checks which registry an NPM_ID_TOKEN was issued for before using it.

Most dangerous docs aren't attacks. They're honest pages telling your agent to fetch something unpinned and run it. We measured how often six search tools pass them on.

As AI coding agents make language syntax less central, runtime performance and operability become increasingly important. This keynote explores why Java’s JVM and standard library are well suited to this shift: newer JDKs continually improve performance, observability, garbage co…

Hi, Wojtek here. Let’s explore this week’s changes in the Rails codebase.

Python 3.15 is out today along with updates for Ruff, uv, and ty.

More on Python 3.15 · Python

Mistral Large 4 is live in Otari from day one. Route to Mistral's 1T-parameter MoE model with one string change, and keep the policies, traces and cost tracking you already run in Otari, Mozilla AI's open-source gateway and policy layer.

Timeline: Large 4 → Model sheet →

More on Large 4 · Large

Trained with reinforcement learning in real environments, Mellum2.1 is built for coding agents and fast sub-agents that run on your own hardware. Today, we’re releasing Mellum2.1, the next version of the 12B mixture-of-experts model we open-sourced in June. The architecture…

Announcing the stable 1.0 release of Genkit Dart, an open-source framework for building AI-powered features and agentic workflows with Dart and Flutter.

More on Dart 1.0 · Dart

Nx 23.3 selects affected tasks instead of whole projects, builds the project graph faster on large repos, and gets ready for Nx 24.

More on Nx 23.3 · Nx

Pydantic v2.14 adds Python 3.15 support, with lazy imports and TypeForm, stabilizes the MISSING sentinel, and speeds up schema generation.

More on Pydantic v2.14 · Pydantic

Claude Dashboards and Claude Motion are in beta. Docs, Slides, and Design are now on every plan, including Free.

Atlas introduces Intelligent Workload Management, improving cluster availability and keeping performance predictable under overload by queuing transient spikes and protecting short-running operations.

Heroku’s continuous platform innovation keeps operational overhead out of your path while delivering our industry-best developer experience. We handle everything below your code, from the core infrastructure and OS maintenance to runtimes and dependencies, so developers can stay…

Prompt caching, first introduced on SambaCloud with MiniMax M2.7, is now available for MiniMax M3. When requests share a stable prefix of at least 4,096 tokens, SambaCloud can serve that prefix from cache instead of recomputing it, with no code changes required. Cached tokens are…