Field Signal · Curated Reading
The Engineering
Signal.
High-signal field notes, architecture teardowns, and engineering writing from 50 publications. A focused reading queue for becoming a stronger end-to-end engineer. Updated 23 Aug.
Accelerating Qwen3-8B Agent on Intel® Core™ Ultra with Depth-Pruned Draft Models
Open the source for the full engineering note.
VibeGame: Exploring Vibe Coding Games
Open the source for the full engineering note.
Nemotron-Personas-Japan: ソブリン AI のための合成データセット
Open the source for the full engineering note.
Partnering with AARP to help keep older adults safe online
OpenAI and AARP are partnering to help older adults stay safe online with new AI training, scam-spotting tools, and nationwide programs through OpenAI Academy and OATS’s Senior Planet initiative.
Swift Transformers Reaches 1.0 – and Looks to the Future
Open the source for the full engineering note.
More ways to work with your team and tools in ChatGPT
New shared projects, smarter connectors, and compliance and security updates help teams get more done.
Measuring the performance of our models on real-world tasks
OpenAI introduces GDPval, a new evaluation that measures model performance on real-world economically valuable tasks across 44 occupations.
Discord Update: September 25, 2025 Changelog
Here's the Discord Changelog from September 25, 2025, so you can stay informed on what’s new in recent app updates!
New Looks for Nitro, New Looks for You. Get Yourself a Nitro-exclusive Profile Bundle!
Between now and September 30th, 2025, new and existing Nitro members can claim a profile bundle matching Nitro's new look, including an Avatar Decoration, Profile Effect, and a Nameplate! Open this blog to see the details.
Introducing ChatGPT Pulse
Today we’re releasing a preview of ChatGPT Pulse to Pro users on mobile. Pulse is a new experience where ChatGPT proactively does research to deliver personalized updates based on your chats, feedback, and connected apps like your calendar.
ENEOS Materials brings ChatGPT Enterprise to manufacturing
ENEOS Materials uses ChatGPT Enterprise to speed research, improve plant design safety, and cut HR analysis time by 90%, with 80% reporting better workflows.
SAP and OpenAI partner to launch sovereign ‘OpenAI for Germany’
SAP and OpenAI launch OpenAI for Germany, a 2026 partnership to bring secure, sovereign AI to Germany’s public sector, enabling safe, efficient public services.
OpenAI, Oracle, and SoftBank expand Stargate with five new AI datacenter sites
OpenAI, Oracle, and SoftBank announce five new Stargate AI datacenter sites, accelerating a $500B, 10-gigawatt U.S. infrastructure buildout to power next-generation AI and create tens of thousands of jobs.
Integrating Grafbase and Redpanda for seamless real-time features
Grafbase launched native support for Kafka through Grafbase Extensions and Redpanda is the ideal streaming engine to power it. Here’s how to integrate them.
Smol2Operator: Post-Training GUI Agents for Computer Use
Open the source for the full engineering note.
CNA is transforming its newsroom with AI
In this Executive Function series from OpenAI, discover how CNA is transforming its newsroom with AI. Editor-in-Chief Walter Fernandez shares insights on AI adoption, culture, and the future of journalism.
SchoolAI builds an AI platform that empowers teachers
SchoolAI uses GPT-4.1, image generation, and TTS to power safe, teacher-guided AI tools for over 1 million classrooms, improving engagement, oversight, and personalized learning.
OpenAI and NVIDIA announce strategic partnership to deploy 10 gigawatts of NVIDIA systems
OpenAI and NVIDIA announce a strategic partnership to deploy 10 gigawatts of AI datacenters powered by NVIDIA systems, with the first phase launching in 2026.
SyGra: The One-Stop Framework for Building Data for LLMs and SLMs
Open the source for the full engineering note.
Gaia2 and ARE: Empowering the community to study agents
Open the source for the full engineering note.
Scaleway on Hugging Face Inference Providers 🔥
Open the source for the full engineering note.
Democratizing AI Safety with RiskRubric.ai
Open the source for the full engineering note.
Detecting and reducing scheming in AI models
Apollo Research and OpenAI developed evaluations for hidden misalignment (“scheming”) and found behaviors consistent with scheming in controlled tests across frontier models. The team shared concrete examples and stress tests of an early method to reduce scheming.
Public AI on Hugging Face Inference Providers 🔥
Open the source for the full engineering note.
Introducing Stargate UK
Open the source for the full engineering note.