Reading room.

How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering?

Recent autonomous machine learning engineering (MLE) agents have made significant progress on public leaderboards. Often motivated by progress stagnation over long-horizon cycles and limited Large Language Model (LLM) pr…

RLTL;DR: Self-Improvement by Internalizing Self-Generated Feedback

The common paradigm of reinforcement learning with verifiable rewards (RLVR) is to let agents make multiple attempts at a task, and optimize towards the successful ones. This becomes problematic in the realms of self-imp…

Announcing AWS Well-Architected Agent, an AI-powered intelligence to optimize your cloud environment (preview)

Introducing the public preview of AWS Well-Architected Agent, an AI-powered service that analyzes your AWS environment to deliver targeted, contextual recommendations for improving your applications' cost, security, perf…

Cauldron Bubble w/o the Trouble? It’s Discord Halloween, Witches

A cold wind whistles, a chill sets in. It’s Discord Halloween: Cauldron Chaos edition! We’ve got frights, we’ve got delights, and we’ve even got some spooky surprises up our sleeve…

10 technical talks I’m excited about at GitHub Universe 2026

From verifying AI-written code to securing npm dependencies, these are the sessions I’m building my Universe agenda around. The post 10 technical talks I’m excited about at GitHub Universe 2026 appeared first on The GitH…

Why I tried to kill token billing (and why we kept it)

Token billing is useful infrastructure, but usually a bad customer-facing pricing model. Your invoice should define the value your product delivers, not break down what it cost you to create it.

On the Effectiveness-Fluency Trade-Off in LLM Conditioning: A Systematic Study

Controlling the output of Large Language Models (LLMs) is a central challenge for their reliable deployment, yet a clear understanding of the involved trade-offs remains elusive. Current approaches to conditioning are of…

SCLATE: A Substrate for Continual-Learning Agent Training and Evaluation

Continual-learning agents are systems of models, harnesses, and memory operating over long multi-session horizons. Evaluating and training them requires interleaving tasks with agent-side events such as session stop and…

Amazon Aurora PostgreSQL now supports direct querying of Apache Iceberg and Parquet data in your data lake

Amazon Aurora PostgreSQL now lets you directly query Apache Iceberg and Parquet data stored in your data lake alongside live operational data—no ETL pipelines required. Powered by DuckDB embedded within Aurora, this capa…

Amazon S3 Tables now support all Apache Iceberg V3 data types

Amazon S3 Tables now support all data types in the Apache Iceberg V3 spec, along with column default values, deletion vectors, and row lineage. Upgrade existing V2 tables or create new V3 tables with built-in compaction,…

Amazon S3 Vectors now supports metadata pre-filtering for higher recall on filtered searches

Amazon S3 Vectors now supports metadata pre filtering, delivering up to 5x higher recall on filtered searches. Filters evaluate before similarity search so scoped queries return more relevant results. Ideal for RAG, agen…

Celebrating Our Newest AWS Heroes – September 2026

Today, we’re excited to introduce the newest members of the AWS Heroes program. AWS Heroes are a vibrant, worldwide group of AWS experts who go above and beyond to share knowledge, mentor others, and build thriving commu…

Seraphina Goldfarb-Tarrant

RCP-nDCG@10: A New Approach to Enterprise Retrieval Quality

Embed 5: Frontier Embedding Models for Enterprise

Gemini 4 Argon: our next era of frontier intelligence

Introducing SynthID Bio

Proof of concept for watermarking AI-generated proteins while preserving biological function.

Forecasting space weather risks on power grids

Extreme space-weather events can damage power systems on Earth and degrade GPS accuracy and satellite operations. A new machine learning system can predict where damage is likely to occur 30-60 minutes before a storm arr…

How Deterministic Controls Turn AI Output Into Reliable Prompt Templates

By Vaibhav Raizada, Kumar Kasimala, and Ashish Gite. In our Engineering Energizers Q A series, we highlight the engineering minds driving innovation across Salesforce. Today, we spotlight Vaibhav Raizada, a senior softwa…

OUSD is now the default stablecoin on Stripe

Open USD (OUSD), a stablecoin built for global money movement, is now available across Stripe. Businesses can use OUSD to manage funds, make payments, and offer new financial services.

Personalization without user identity

How Airbnb uses proximity signals to personalize without relying on individual user history. By : Wei Jiang , Bin Xu , Bharathi Thangamani , Weiwei Guo , Sundar Srinivasavaradhan , Tracy Yu , Huiji Gao , Swapnil Ghike ,…

The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models

When language models reason in chain-of-thought or exchange free-text intermediates, they serialize structured information into natural language. How much tree-structured compositional content survives this bottleneck? W…

Evolving our calendar assistant Reclaim to be AI-native without starting over

How we redesigned Reclaim for AI and natural-language requests while preserving the scheduling experience people rely on.

That Which We CAN Do: Sustainability in a World of AI

As computing use grows, so does its environmental footprint. While hardware and software have become significantly more efficient, AI demand is growing even faster, outpacing those efficiency gains and increasing the ove…

Developer policy update: Transparency, state policy, and what’s ahead

Explore GitHub’s latest transparency data and learn more about policy updates affecting developers and open source. The post Developer policy update: Transparency, state policy, and what’s ahead appeared first on The Git…

GitLab and Claude Code: Fast, compliant AI

Government agencies are feeling twin pressures: While the U.S. Office of Management and Budget (OMB) is urging you to deploy AI faster, the U.S. Government Accountability Office (GAO) wants guardrails in place before tha…

Build with Azure Canvases: a shared workspace for agents

See how Azure Canvases bring plans, code, previews, and deployment feedback together in a collaborative workspace for agent-assisted development. The post Build with Azure Canvases: a shared workspace for agents appeared…

Enabling Consistent AI-Assisted Engineering with GitHub Copilot Plugins

When an engineer or team creates useful agentic artifacts for their daily work (e.g., instructions, skills, and MCP configurations), others naturally want to adopt them when they see how those artifacts could help with t…

Introducing Quine: An AI research system designed for the complexity of biology

Biology doesn't operate in silos, and neither should the AI representation of it. Quine is an early-stage research effort to create a multimodal world model of biology. By connecting insights across biological scales and…

Helping personal agents shop more intelligently and reliably with Link

As agents take on more purchases, agent builders have increasingly asked us to help agents navigate checkout and earn consumer trust. Today, we’re sharing three major improvements to support this broader set of needs.

The making of the Yelp Assistant UI

As simple as it may look, a chat application is made up of thousands of small decisions that we make along the way, all with the goal to build a delightful experience for users. But each decision comes with a cost, often…

Faster Rates for Federated Variational Inequalities

In this paper, we study federated optimization for solving stochastic variational inequalities (VIs), a problem that has attracted growing attention in recent years. Despite substantial progress, a significant gap remain…

AWS Weekly Roundup: GPT-6 Sol and Luna, Claude Opus 5.5 on Amazon Bedrock, Strands harness, and more (September 28, 2026)

If there’s one theme that defined last week, it’s choice. The frontier models keep arriving, and the interesting question is no longer just “how smart is it?” but “which model fits this step, at this cost, at this latenc…

Highlights from Git 2.56

The open source Git project just released Git 2.56. Here is GitHub's look at some of the most interesting features and changes introduced since last time. The post Highlights from Git 2.56 appeared first on The GitHub Bl…

How we found 24 Android vulnerabilities using our open source AI security agent

A look at the targeted AI taskflows behind these findings, the critical Android bugs they uncovered, and how to run the same open-source agent on your own app. The post How we found 24 Android vulnerabilities using our o…

What's new in Git 2.56.0?

The Git project recently released Git 2.56.0 . Let s look at some of the highlights of the release, including contributions from the Git team at GitLab. What s covered: Git Merge 2026 and schedule for Git 3.0 git-history…

Get started with the GitHub Copilot app: a free, hands-on course

Learn to direct AI coding agents, review their work, and build reliable development workflows with this free GitHub Copilot app course. The post Get started with the GitHub Copilot app: a free, hands-on course appeared f…

Mistral Opens Munich Hub to Advance Industrial AI in Germany

One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact

Since launching a year ago, the Microsoft Research Asia — Singapore lab has established a strong foundation, deepened collaboration across government, academia, and industry, and explored how frontier AI research can cre…

Engineering Multi-Agent AI Teams That Build and Test Themselves

By Sohini Arya and Manish Kumar Jha. In our Engineering Energizers Q A series, we highlight the engineering minds driving innovation across Salesforce. Today, we spotlight Sohini Arya, AI Delivery Lead for Marketing Clou…

Travel’s AI dilemma at Skift Global Forum

At the “great recalibration”-themed travel conference, Airbnb CEO Brian Chesky called AI “an existential risk” to his company and “literally the best thing to ever happen” to it in the same discussion. That sums up many…

2028: Two scenarios for global AI leadership

Compass is coming to the cloud

Discord Update: September 25, 2026

Here s the Discord Changelog from September 25, 2026, so you can stay informed on what’s new in recent app updates!

GitHub Copilot app for Beginners: How to build custom workflows with canvases

Describe the interface you need in plain English, then let the agent build a live surface you can both use and update—so you spend less time adapting to tools and more time getting work done. The post GitHub Copilot app…

Improving site performance by shipping more CSS

How we fully migrated github.com away from CSS-in-JS. The post Improving site performance by shipping more CSS appeared first on The GitHub Blog .

Automated merchant review summary system with integrated feedback

Introduction Merchant reviews contain useful details about food quality, portion size, packaging, and value. Finding these details often requires reading many comments. Aggregate ratings simplify comparisons, but they do…

Trading a Cloud Identity for Your Own: Workload Attestation on Managed Compute

By Dhruv Pratap Introduction Organizations that have been around for a while usually run two identity systems side by side. One belongs to the cloud provider: IAM roles, instance profiles, execution roles. The other is y…

Partition Finalization in Pinterest’s Next-Generation DB Ingestion Framework

Qianrui Zhang | Sr Software Engineer, Logging Platform Kanchi Masalia | Software Engineer II, Stream Processing Platform Liang Mou | Sr Staff Software Engineer, Logging Platform Yi Pan | Principal Engineer, Agent Platfor…

Compressing Streaming Neural Audio Encoders via Latent-Space Distillation

System-wide Dictation on Apple devices runs entirely on-device, and the speech it transcribes reaches the foundation model through a tokenizer: an encoder that maps short windows of waveform onto the representation the l…