Reading room.
Engineering blogs & AI research
A reading room from engineering teams and AI labs: titles, short feed excerpts, and links to the originals. For my own essays, explore the writing archive.
- 1359 articles tracked
- 37 sources
- Refresh scheduled daily
Sources
FERN: Cross-Benchmark Table Retrieval for Text-to-SQL with Frozen Retrievers and Learned Ranking
Text-to-SQL systems over large pooled catalogs must first retrieve the tables a question needs. We compare against the published table recall of RASL, which uses Claude 3.5 models to extract keywords and select tables, o…
Pterodactyl: Pushing the Limits of Efficient Automated Translation Evaluation
In this paper, we present 1020 Group's submission to the WMT26 shared task on automated quality evaluation. Our system, Pterodactyl, was engineered specifically for subtasks 2 and 3, that is, segment-level quality score…
Mechanistic Interpretability Needs a Logic of Inference
Neural networks are increasingly studied not merely as predictive systems, but as complex systems whose internal computations are themselves objects of scientific inquiry. Recent progress in mechanistic interpretability…
Comparative Analysis of Agent Evaluation Frameworks: Stability, Detection, and Discovery in Enterprise Analytics Agents
Deploying LLM-based analytics agents in enterprise settings requires evaluation frameworks that can reliably detect failures across complex, multi-tool workflows. We present a three-phase comparative study of three evalu…
Contrastive Discovery: Open-Ended Scientific Discovery over Competing Explanations
LLM agents are increasingly used for open-ended, data-driven scientific discovery, yet most of them evaluate hypotheses pointwise, testing an empirical claim $h$ only against its complement $\neg h$. When evidence reject…
AutoScaffold: AI-Guided Scaffolding for Reinforcement Learning of Language Models
Reinforcement learning with verifiable rewards (RLVR) produces rich training experience, yet outcome rewards reduce each attempt to a single number and leave most of this information unused. Training-time scaffolds, such…
Gemini 4 Carbon is now available
Introducing AlphaProof Code: a breakthrough in verifying complex software at depth
AlphaProof Code agents wrote three million lines of proofs in Lean to verify Signal’s core encryption.
Git security vulnerabilities announced
Git is releasing updates addressing six security vulnerabilities affecting recursive clones, reference handling, merges, binary patches, and imports. Here is what users need to know. The post Git security vulnerabilities…
Decision Models in Hugging Face
Pay for Hugging Face Hub with your Google Cloud Account
Beyond Test Cards: A Fake Payment Provider for Testing at Scale
Image generated with Google Gemini, 2026 Written by Savelii Kolesnikov , Jay Patel , and Ivanna Sivak Background Lyft processes billions in payments annually, and that volume continues to grow. Agentic coding has drastic…
Hiring Junior Engineers
Several people have asked me what the qualifications are for junior software engineers to be hired these days. This level is often referred to as “new grad,” but people with lots of different backgrounds, not just recent…
Your Algorithm: Giving People More Ways to Personalize Facebook Recommendations
Your algorithm is a suite of AI-powered tools that makes Facebook recommendations more transparent and steerable, with a single editable view of the interests that shape Feed and Reels. We’re sharing how Your algorithm t…
How We Built TLX: A Hardware-Native GPU Compiler for Production ML at Scale
Triton Low-Level Extensions (TLX), is an extension to Triton that helps engineers optimize GPU performance while preserving Triton s productive programming model. TLX introduces explicit warp-group and cluster-level orch…
Lazare, veni foras: Reprogramming Dormant Neurons to Backdoor AI Agents
Microsoft researchers have discovered Lazarus, a backdoor technique that can reprogram dormant neurons in an AI model, causing an otherwise normal AI agent to execute malicious commands when a hidden trigger appears. The…
Agent Substrate: Advancing Kubernetes for Agents at Scale
AI agents spend most of their time idle, waiting on a human prompt, a tool result or a model response. Kubernetes wasn’t built for that. Running one Pod per...
Manage Kubernetes Clusters for GPU Infrastructure with NVIDIA Kubernetes Engine
NVIDIA is announcing two open source projects for building and operating managed Kubernetes services on GPU infrastructure—NVIDIA Kubernetes Engine (NVKE) and...
Running AV Reasoning Models Within Onboard Compute and Latency Limits
Autonomous vehicle (AV) developers need reasoning models that can operate within the compute and latency constraints of an onboard system. Larger models provide...
Start and Restore GPU Workloads in Seconds with NVIDIA Snapshot
NVIDIA Snapshot, a new open-source component of the NVIDIA DSX AI Factory Platform, makes initialized GPU workloads reusable across inference...
Work at the Frontier: AI’s effect on freelance work
New research from OpenAI and Upwork finds shifting demand for freelance work, new skill combinations, and larger budgets and more hours for AI-related projects.
Closing the AI guidance gap in education
New ThinkYoung research and lessons from OpenAI’s EMEA education partnerships on supporting teachers, AI literacy, and independent learning.
Cogent and OpenAI build cyber defense factories for enterprises
Cogent combines OpenAI models with enterprise context to help automatically remediate vulnerabilities with agents.
Helping caregivers use AI in everyday life
OpenAI Academy is bringing free, hands-on ChatGPT workshops to caregivers in five U.S. cities, helping them build practical AI skills for everyday life.
Introducing Businesses in ChatGPT
Manage your business information, get insights about how your business shows up, and turn interest into action in ChatGPT.
Singapore Airlines elevates travel & work with OpenAI technology
Discover how Singapore Airlines uses OpenAI models and API to enhance customer support, travel discovery, and employee productivity, doubling CSAT for its Kris chatbot.
How Salesforce Scaled Engineering AI Adoption Without Stifling Innovation
By Paul Constantinides, Anubhav Dhoot and Jackie Chambers de Freitas Salesforce s Hyperforce Runtime Platform (HRP) engineers were already experimenting with AI. But promising capabilities remained scattered across teams…
AWS Weekly Roundup: AgentCore Runtime V2, Amazon DynamoDB bulk put items, and more (November 9, 2026)
Most of the cost of adopting something new isn’t the new thing, it’s the move. Moving a Hive table to Apache Iceberg has typically meant weeks of engineering work. Last week, Amazon S3 Tables turned that into three steps…
BAIR at CoRL 2026
The Conference on Robot Learning (CoRL) takes place November 9–12, 2026, in Austin, Texas. BAIR faculty, postdocs and students will be presenting research at various sessions, including an Early Career Keynote from Profe…
Launching the Vite Ecosystem Fund
Cloudflare is distributing the first $301,300 of the Vite Ecosystem Fund to open source projects and maintainers. The recipients were chosen by a panel of Vite stack core team members, including maintainers not employed…
Gemini Robotics 2.5 accelerates real-world general dexterity
Our most capable vision-language-action model enhances robot dexterity, reliability, and safety
Discord Patch Notes: November 9, 2026
Check out the finer details of the more technical fixes implemented into Discord recently.
Unlock, Level Up, Flex: Discord Profile Badges Got an Upgrade
Meet the new badge directory: one spot to browse every badge on Discord, including 4 shiny new ones to unlock, level up, and flex.
How to design a decision-grade pilot for agentic AI
Agentic AI changes the nature of a technology evaluation. When an AI system only generates an answer, a person reviews the output and decides whether to use it. An agent goes further: It gathers context, selects tools, m…
Your Experts Have Partners. There's HOPE for Keeping Them Together
Nemotron-Personas-Finance-Japan: A Synthetic Dataset Capturing Japan's Demographics and Financial Context
Extending LLDB to Debug Applications That Span CPUs and GPUs
We built a vendor-neutral accelerator plugin architecture in lldb-server for debugging applications that span CPUs and GPUs. It separates shared debugger infrastructure from hardware-specific device control. The design r…
Do AI coding agents care about SemVer?
We gave AI coding agents the same breaking upgrade twice. Labeled 2.0.0, Claude Sonnet 5.5 checked it in all 10 runs. Labeled 1.4.3, it checked in two and often left the app broken. Nearly 1,000 runs show which agents st…
Fact-Checking at Scale: Building a More Accurate and Efficient Synopsis Factuality System
By Shreya S Ramanujam , Yongce Li , Afshine Amidi , Sina Ghiassian , Benjamin Klein . Introduction Picture this — you’re home after a long day at work and can’t wait to catch up on Bridgerton. You log on to Netflix, sear…
Make a Robot Arm Feel Weightless with NVIDIA Holoscan 5
Picture guiding a robot arm by hand while its motors supply just enough torque to support its weight and keep the robot upright. This simple interaction, a...
ChatGPT Enterprise now available in AWS Marketplace
ChatGPT Enterprise is now available through AWS Marketplace, helping organizations simplify procurement through existing AWS purchasing and billing relationships.
CodeRabbit scales agentic software development and ships 20x more
CodeRabbit uses OpenAI models to govern and trust AI-generated code and Codex to ship 20 times more software with the same engineering team.
Control Room: A central system to track and stop model workloads across our research infrastructure
Learn how OpenAI’s Control Room tracks model workloads across our research infrastructure and can rapidly stop them when safety or security concerns arise.
Kuken Industries searches 70 years of drawings in minutes
Kuken Industries uses the OpenAI API to search 70 years of cooling tower drawings. Engineers can find the drawing for a repair in 4.5 minutes, down from nearly an hour.
Using AI to Build AI: How Agentforce Engineers Scale Automation Without Sacrificing Reliability
In our Engineering Energizers Q A series, we highlight the engineering minds driving innovation across Salesforce. Today, we spotlight Fereydoon Vafaei, a Principal Software Engineer on the Agentforce team, who faced a f…
AI Product Management at Spotify: A View From Three Angles
Almost every article or podcast about product management and AI refers to "the AI PM" as one job. But what... The post AI Product Management at Spotify: A View From Three Angles appeared first on Spotify Engineering .
Zero-Copy Migration: How Yelp Moved Its Cassandra Fleet Between Kubernetes Clusters
Introduction How do you move a large Cassandra fleet of around 1000 Kubernetes Pods and a petabyte of data across Kubernetes clusters without anyone noticing? Apache Cassandra powers many of Yelp’s core features, from us…
The post-quantum future of web privacy
Cryptographic protocols like Privacy Pass help keep web sites safe from abuse while keeping their users private. This post looks at the quantum threat to these protocols, and some new post-quantum cryptography that may h…
Beyond code generation: A developer’s guide to agentic engineering
Writing code got faster. The work around it didn't. See how you can plan, write code, review changes, and investigate production issues with GitHub Copilot, and what work still needs your judgment. The post Beyond code g…
The Pentagon wants to know who checks code when AI writes it
The Department of War now treats AI-generated code as unverified until it’s proven otherwise. DoW Instruction 8430.01 , “Accelerated Mission Software,” took effect September 8, 2026, and is the department’s latest guidan…