Tim J. Brauer·6d agoThe Misunderstood Genius: A Guide to Leveraging AI in the Long RunHave you ever felt less productive because of AI agents? You jump between five chats, check intermediate results, give feedback, and at the…
Tim J. Brauer·Sep 30Ambidex — Autonomous research loop combining exploitation (autoresearch-style) with explorationA run on Tiny Shakespeare. Each line is one research direction the agent found while exploring and then pushed in exploitation.
Marius·Sep 13Out of the chatIn short: AI is already good enough to work for hours without you. What’s missing is the infrastructure around it. Control over what runs…
Floris Fok·Aug 19Applied Auto Research: My layer problem solved by AI.We got more juice out of a fast cheap model by leveraging auto research. Not “can an agent do novel science,” but “how small a problem is…
Rahultiwari·Aug 18AutoResearch: When the Experiment Loop Becomes the ResearcherThere is a part of machine-learning work that sounds exciting when we describe it and feels very different when we actually do it.
anvilarth·Aug 15Age of AutoresearchExpertise is moving from “find the solution” to “build the system that finds it”. Notes from a month and a half of doing it wrong, then…
Mohan Paramasivam·Aug 12Karpathy’s “Autoresearch” Just Went Viral — Here’s How Software Engineers Can Actually Use the…Forget neural networks for a second. The real idea inside this repo is a blueprint for letting AI agents run unattended overnight — and it…
InFlexport EngineeringbyFlexport·Aug 4We built an AI that does AI research on another AIA few months ago Andrej Karpathy released autoresearch, an open-source pattern for letting an AI run experiments on machine learning code…
Andrei Matsko·Jul 30Creating an agent is not enough — we need to test the testerEvals for QA Pipeline and auto-improvement loop
InBetter MLbyJaideep Ray·Jul 25What do Self-Improving Agents learn from their mistakes ?Darwin Gödel Machines
