What's moving in quantum programming, software engineering and agentic engineering.
Straight from the people building it.
My obituary for programming Jeremy Scheff poignantly captures the key loss caused by coding agents for career programmers who enjoyed the puzzle of developing code: Programming turned into one of the greatest jobs you could imagine. You got paid to work on tricky little logic puzzles. The trickier the puzzle, the more you got paid. But even pretty simple puz
Sdists rebuilt from main for /update on the CLAI bleeding channel. Not a release.
Our first scanner read config files and missed what was running. We rebuilt it with the cloud inventory, a careful probe, and honest results.
This week's Java roundup for October 12th, 2026, features news highlighting: three OpenJDK JEPs targeted for JDK 28; the October 2026 edition of Azul Payara; point releases of Gatherers4j, Yupiik Fusion, Google ADK for Java and Google ADK for Kotlin; and 38 CVEs resolved in Apache Camel. By Michael Redlich
Learn how to move an agent session from the GitHub Copilot app into VS Code without losing the conversation, code changes, or working context. The post How to continue a GitHub Copilot app session in VS Code appeared first on The GitHub Blog.
We are updating the Terms of Service for Hobby and Pro customers, effective October 19, 2026. The updates clarify some existing Vercel practices, expand terms on topics such as managed accounts and credits, and reflect applicable legal requirements, including certain cancellation and service-change rights for eligible consumers in the UK and EU. A new sectio
Apple and WebKit have shipped protections against web tracking for a long time.
Use an agent skill to prepare your existing .NET MAUI app for Xcode 27, with the scene lifecycle changes and manual migration steps explained. The post Get Your .NET MAUI App Ready for Xcode 27 appeared first on .NET Blog.
Generative Recommenders (GenRecs) represent a paradigm shift from scoring every ad to reasoning about ad content and user preferences together in order to predict the best outcome for advertisers. Leveraging the world knowledge built into LLMs makes ad matching more intelligent and precise, compounding gains for advertisers. GenRecs Tokenization is the found
The GitHub Copilot SDK gives your application an agent runtime. BYOK lets you keep that runtime while sending model requests to your own Microsoft Foundry or Azure OpenAI deployment. Here is the complete C# setup. The post Bring Your Own Model to the GitHub Copilot SDK with C# appeared first on .NET Blog.
A maturity model for taking agents from an impressive demo to a system people can depend on — with a self-assessment and the map to a deep-dive on each layer.
Sometimes, giving an AI agent all of the knowledge it needs is the wrong approach. At Meta, we’ve seen notable improvements in efficiency and accuracy by keeping the orchestration layer of our data pipeline agents deliberately thin: It routes work and enforces reliability checks, while specialized agents do the domain reasoning. We’re sharing our own [...] R
Pinterest has introduced event time-based partition finalization for its database ingestion framework. The mechanism uses Apache Flink to track event time and Apache Iceberg metadata to publish a non-regressing watermark, helping downstream jobs determine when continuously ingested CDC data is sufficiently complete for processing. By Leela Kumili
JDK 28 finalizes structured concurrency through JEP 543, with no API changes from JDK 27. Teams on JDK 27 only need to drop --enable-preview and recompile. Code written for JDK 25 or 26 must adapt to the R_X exception type parameter, ExecutionException wrapping, and the abstract timeout() joiner callback. By A N M Bazlur Rahman
With the recent release of the Muse family of foundational models, we want to take you inside the infrastructure that has made it all possible. The first phase of Prometheus, our gigawatt-scale supercluster, is already complete and we’re working toward scaling it to comprise ~1 million GPUs. Prometheus also features technologies like closed-loop liquid cooli
The article examines the transition from microservices to agentic AI architecture in distributed systems. Microservices allow for functional decomposition, while agentic systems focus on decision decomposition. This shift introduces new complexities in observability, reliability, and system design, highlighting the need for updated architectural patterns and
Rachel Lee Nabors, former React core team member, dives into the architecture and philosophy behind local-first applications, exploring how small language models (SLMs) running directly on-device can deliver privacy-focused, low-latency, and cost-effective alternatives to cloud-hosted LLM wrappers. By Rachel Lee Nabors
Based on Gemma 4 and released under the Apache 2.0 license, Google's EmbeddingGemma 2 is a compact, open-weight, multimodal embedding model designed to make semantic search and retrieval-augmented generation (RAG) more efficient while optimizing performance for on-device deployment. By Sergio De Simone
Continuous integration provider Buildkite has introduced pipeline-scoped secrets and new Model Context Protocol tools for validating pipelines, diagnosing failed builds and repairing webhooks. The company has also added Linux arm64 support to its GitHub Actions compatibility layer and expanded OpenTelemetry data in Buildkite Agent v4.1.0. By Matt Saunders
Microsoft has released .NET 11 Release Candidate 2 under a go-live license, supported in Visual Studio 2026 Insiders. The release adds incremental CBOR reading, experimental post-quantum cryptography APIs, file-based app directive improvements, NuGet package cooldowns, F# runtime async building blocks, a C# UI option for .NET MAUI, and refined Windows Forms
NestJS version 12 has been released, transitioning core packages to ESM and adding Standard Schema validation. It updates the CLI and introduces an observability SDK. Key changes include project generation choices between CJS and ESM, improved validation options, and lifecycle project hooks. Users will need Node.js 20.19+ or 22.12+. Migration guides are avai
Cheaper code won't kill software development, but it’s already reshaping it
When code becomes abundant trust becomes scarce. That is the argument GitLab CEO Bill Staples made in his recent essay: AI is collapsing the cost of producing software, and everything around production, including context, verification, governance, and evidence, becomes proportionally more important. Both security and collaboration are key elements to produci
is now available on AI Gateway with 50% off launch pricing. Inception reports strong results across a range of benchmarks, with generation speeds above 1,000 tokens per second.Mercury 3 from Inception Mercury 3 uses diffusion to generate text through parallel refinement. It brings improved reasoning to latency-sensitive workloads such as browser agents, real
from is now available on for free until Oct 22.MiniMax M3.1 FlashMiniMaxAI Gateway MiniMax M3.1 Flash is an open-weight multimodal model for coding and agentic work, with a 1M-token context window for long documents and multi-step agent sessions. Compared with earlier MiniMax models, M3.1 Flash improves on complex agentic tasks, coding, and cybersecurity. On
You already have an AI assistant. It completes your code, explains code you didn’t write, and maybe even drafts a whole feature while you sip your coffee. An agent on GitLab picks up where your editor stops. The AI in your editor only knows what's open in your editor. It doesn't know that your merge request has been sitting in the queue for two days, or that
now have a that groups requests into sessions and shows how session affinity routed each one. Each row shows the or app that ran the session, such as Claude Code, Codex, or Cursor, along with its cost, models, providers, and how much of its input was cached across every turn.AI Gateway logsSessions viewcoding agent Open a session to see its cache hit rate, d
Google has replaced its long-standing Python type checker, Pytype, with Pyrefly, an open-source tool originally developed by Meta. The decision was driven by Pyrefly's advantages, including faster and more accurate type checking, and lower infrastructure costs. By Sergio De Simone
AWS recently announced the preview of the Well-Architected Agent, a service that analyzes environments against Well-Architected best practices and provides prioritized recommendations across cost, security, performance, and reliability. The AI agent can also suggest remediation steps and review infrastructure-as-code before deployment. By Renato Losio
On October 24, AWS will operate its European Sovereign Cloud without a connection to the AWS Global Network backbone, with a team of EU residents and the region’s own hardware and software. AWS says neither of the two systems being severed is operation-critical, and practitioners are divided over whether the exercise answers the question customers are asking
Reverse Engineer Anything (REA) is an open-source project with over 200K GitHub stars that connects coding agents such as Claude Code and Codex to reverse-engineering tools like Ghidra, Hopper, and IDA. The project lets developers inspect compiled binaries, trace function calls, and investigate how applications implement specific features. By Renato Losio
Amazon has acquired DuckLabs, the company behind the analytical database DuckDB. Co-founders Hannes Mühleisen and Mark Raasveldt will continue to lead the team under AWS. The acquisition aims to enhance the development and reach of DuckDB while maintaining its open-source status. Amazon Aurora PostgreSQL gets its first DuckDB-powered feature following the ac
is now available on .DeepSeek V4.1 Flash 1017AI Gateway DeepSeek V4.1 Flash 1017 is a minor update to that improves reasoning, coding, agentic tasks, and image understanding.DeepSeek V4.1 Flash The coding gains show up in terminal work and in building and editing whole repositories, and the agentic gains in multi-step automation. With images, the model reads
Google researchers open sourced EnvHarness, a framework that dynamically adapts agent-training environments to an agent's weaknesses without modifying the agent, environment, or evaluation system. According to the researchers, this approach can improve both learning effectiveness and efficiency while helping agents overcome their weaknesses. By Sergio De Sim
Zod is a TypeScript-focused schema validation library that recently released versions 4.5 and 4.6. Key updates include schema compilation for faster data parsing, significant memory usage reductions, and new validation features. The library supports efficient validation processes and remains popular in the JavaScript ecosystem, achieving millions of download
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.
Changelog Added Define Custom Operations via the C API (#16549) Add base implementation of LoweringPassManager (#17105) Define Pass and PassContext handles for Python (#17104) Add implementation of IR handle extraction (#17103) Define Python API for Rust-native PassManager (#17089) Add Delay with units of Dt to the C API (#16915) Expose the pass manager to C
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 and model-based testing can verify AI-generated code against real-world productio
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 agents to shop. Today, , our open source storefront template for , is generally
Microsoft will scale its Maia accelerator family to multi-gigawatt deployments from 2027, making it the primary inference system for its own products. Maia 300 collapses scale-up and scale-out onto one Ethernet fabric and moves KV state between prefill and decode workers without leaving it. Practitioners have questioned the tokens-per-watt claim that leads t
Nick Cammarata out of nowhere with some major claims!
What's Changed ⚠️ Compatibility Notes Attach namespaced origin and tool_call_id metadata to BackgroundTools completions by @michaeldunn9 in #10196 Take GPT-Live realtime history, usage, and reply waits from the session core by @DouweM in #10083 Honour disable-model-invocation and user-invocable in Skills, and skip stray skill library paths by @DouweM in #103
, previously available in stealth as , is now available on . Beam is free to use for a limited time during the launch period when serviced by Reflection AI.Beam from Reflection AIAI Gatewaystealth/cat-jam Beam is an low latency US open-weight reasoning model for coding and agentic tasks, with . Beam supports and , with a 1M-token context window for code, doc
The Go Blog Interactive refactoring in gopls pre:nth-of-type(1) { background: var(--color-diff-old); } #content .beforeafter > pre:nth-of-type(2) { background: var(--color-diff-new); } .beforeafter-arrow { place-self: center; margin: -0.5em; } .beforeafter-arrow::before { content: "⟶"; @media screen and (max-width: 57.7rem) { content: "⇓"; } } #blog #content
Why executive judgment decides who wins with AI
and are now available on . Both are multimodal for classifying content, routing requests, and scoring text, structured records, and images.Perplexity Decider V1.2 27BDecider Flash V1.2 4BAI Gatewaydecision models The models answer typed questions with probabilities. Boolean questions return the likelihood of yes, Choice questions select from categories you d
Two engineers shipped Stack Overflow for Agents in a few weeks with Codex, and its ChatGPT plugin helped drive up to 10× growth in agent creation.
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 from Elorian Glimpse can return to visual evidence as it works: zooming into a detail, revisiting a short interval of video, or tracking the same feat
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.
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 took to build Forum with agentic AI: how we [...] Read More... The post How Meta Built Forum W
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 none are bigger than the quadrennial [...] Read More... The post 2026 World Cup:
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.
Most engineering teams capable of building an internal AI knowledge system think they should. But six months later, it's a platform team's full-time job. Building an in-house AI context layer costs more money and energy than you think.
Also: insider details on how and why Anthropic moved off Git, a new trend of AI-assisted coding rounds inside of Big Tech, and more
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.
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 recognition celebrates a decade of impact from work that began at University College L
Instagram Direct migrated its Android UI from the legacy Android View system to Jetpack Compose, creating an AI-native UI codebase that is easier for AI coding agents to understand, modify, and maintain. According to Instagram engineers, the migration reduced UI code volume by 50% and token costs per AI agent session by 33%. By Sergio De Simone
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 Francisco to do just that at Vercel Ship 2026. Guillermo Rauch (CEO, Vercel) opene
When software meets hardware, engineering principles remain largely the same, but the sensible defaults often change. Kuldeep Singh describes our experience modernizing a large IoT fleet of battery-powered sensors. Our familiar principles remained important; what changed was how we applied them. These practices emerged through experiments, failures, and oper
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 foundation remains familiar: typed record definitions, stores and […] The post Vect
Software only has value when shipped; faster release cadence matters. Development, deployment, and operations require different thinking, metrics, and investments. There are four impacts in tech: product success, compute capacity, engineering capacity, and strategic innovation. The biggest expense is people. AI shifts engineers toward deeper problem understa
Brendan O'Leary discusses how real-world AI adoption depends on building trust rather than chasing model benchmarks. Drawing from 40 trillion tokens of usage data, he explains the four rungs of the AI adoption ladder and analyzes key failure points in context construction, multi-model routing, and feedback loops to help engineering leaders successfully scale
Grafana Labs has released Tempo 3.1, the first minor update to the 3.0 architecture. It adds TLS and additional SASL mechanisms to the Kafka client, lets operators redact traces with a TraceQL query instead of a list of trace IDs, and introduces an experimental query-time hint that corrects TraceQL metrics for sampled data. By Claudio Masolo
Spring Authorization Server now lives inside Spring Security 7 under the same Maven coordinates, so the upgrade is a version bump. This server-side guide covers enabling DPoP and RFC 8693 Token Exchange, explains why 7.1 apps can silently stop enforcing DPoP until they call .dPoP(), and shows how to combine both into narrowed, sender-constrained tokens for s
AI Gateway Production Index — October 2026 Every month, 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 Production Index reports from , , and .AI GatewayJulyAugustSeptember In August, open-weight mod
Last year, we expanded for Python and TypeScript backend frameworks. By this summer, developers could deploy their frontend and backend together with and coordinate background work with and .zero-config supportVercel Servicesqueuesdurable workflows Today at , we are expanding the backend workloads you can run on Vercel. Python on Vercel is now generally avai
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 manages concurrent requests and scales with traffic. Idiomatic Python SDKs let you wo
is now in public beta, allowing you to run long-running processes that pull and execute work from , databases, or third-party systems without requiring an HTTP server:Vercel DaemonsVercel Queues Daemons run containers on Fluid compute and deploy alongside your app in the same project. Vercel builds your image and maintains the instance count you configure in
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 that from August 2025 to June 2026, non-developer Codex users grew 189x. OpenAI recently reported Our customers call these builders citizen developers. They know their part o
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. Token Pass runs through , so you can choose your tools and switch between inclu
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 to keep an agent’s files independent of its execution environment. It is now generally available for
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 from Meta Muse Spark Open supports text and image input with a 1M-token context window, allowing agents to work with code and screenshots in the same reque
Renting the Frontier, Owning the Core IP: How We Cut AI Forecast Costs from $40M to $11M
Evaluating decision models for document categorization
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
AI agents have changed how much code a team can produce in a day. Review queues haven’t kept up, and delivery stalls while changes wait for a closer look. The usual workarounds (reminders, reviewer rotations, assigning a team) buy a little time and then burn people out. Client-side review with AI helps the author by catching issues as the code is being writt
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 Vercel handles the infrastructure so you can focus on your application: Within your application, Python SDKs let you build agent
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 one of those changes has to clear review before it reaches users. Automated review
Enterprise teams can now run Vercel Functions in their own AWS account with (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.Bring Your Own Cloud During deployment, you provision functions in your AWS account and Vercel serve
now supports in public beta. Your apps and agents can access supported services on behalf of signed-in users without separate consent prompts for each service.Vercel ConnectEnterprise-Managed Authorization (EMA) Here's how access works: If your app uses , you can request a token for with the signed-in user’s identity:Vercel PassportSlack’s Model Context Prot
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 stable If you don't have it already, you can get rustup from the appropriate page on
is now available on as a stealth model, free to use during the stealth period.Cat JamAI Gateway Cat Jam is a reasoning model for coding and agentic tasks, with low-latency inference. It supports , so agents can connect it to your own tools, and that follow a JSON schema you supply. Responses include the model's reasoning text alongside the answer. Cat Jam ta
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.
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.d1-ultra from Liquid AI Rather than generating text, d1-ultra returns structured answers with probabilities. Ask a Boolean question to get the
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 not. Standing up a standalone vault narrows that exposure but requires mirroring ever
now supports , allowing connectors to reach providers with IP allowlists, such as Salesforce and Snowflake.Vercel ConnectSecure Compute With Secure Compute routing enabled, Connect’s provider requests use the selected network’s dedicated static IP addresses instead of Connect’s shared IPs. Requests to obtain and refresh access tokens follow that route; your
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.Vercel Connect Connect creates subscriptions on the MCP server and renews them automatically as needed, verif
The shift to the AI-first enterprise: Insights from technology leaders forum
The future of SaaS in the agentic world: Beyond the disruption narrative
is now generally available for on Hobby, Pro, and Enterprise.Vercel DrivesVercel Sandbox 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 mounting the same Drive. Use Drives to preserve an agent’s workspace or file-based me
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 and use up their context window on extra calls. The GitLab MCP server, now generally
is now available on .Jev 1.14 from TypeSafe AIAI Gateway Jev is a 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, in parallel, for classification, routing, scoring, and automated verification.decision model J
At AI Engineer New York, frontier labs and financial firms explored how to give agents useful autonomy while keeping consequential decisions under control.
How Leash Bio tamed DELs, the assay that traumatizes everyone, and used the data to catch cheating models. And why that same data is perfect for AI agents learning to think like chemists.
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.
QuantumDev started with my morning reading: dozens of publications, then Google Scholar. I built this free digest to help more people discover what is being built, and spend more time learning.
Alexandre Pinel · Creator of QuantumDev