autoresearch

A tweet from @_philschmid speculates that future AI agent harnesses will focus on coded extensions for automatic integration, with Pi leading the trend and DeepSeek as an extreme example, all driven by autoresearch and recursive self-improvements.

This paper identifies a scalability bottleneck in RL-trained automatic research agents—environment execution dominates training cost—and proposes World Model RL (WMRL) with online debiasing and inverse-variance denoising to replace real execution, achieving 3–4x training speedups and better generalization.

This paper identifies common failure modes in tree-search-based autoresearch agents applied to tabular datasets, such as repeated bug resolution, poor hyperparameter tuning, and ineffective exploration, and proposes targeted interventions like a global debug consultant and refined tree-search algorithms to recover wasted compute and improve performance without changing the underlying language model.

User iamai_omni praises Fable 5's self-correction ability, considering it comparable to ASI. Citing a recommendation from yibie, they point out that the Superpowers author had Fable 5 run an autoresearch loop, spending $165 to complete 25 experiments, increasing build speed by 50% and reducing token overhead by 60%, and documented failures and correction processes in detail.

The article describes an experiment using Claude Code to autonomously develop a file compression algorithm with constrained optimization, evaluating the viability of AI agents for unsupervised problem-solving.

Introspection, a new AI startup founded by ex-xAI engineers, introduces 'autoresearch' – a feedback loop system where agents maintain and improve themselves using signals, evals, and human input, moving beyond traditional agent harnesses.

A Twitter thread highlights emerging autoresearch labs that are building AI systems to automate the full research loop, from hypothesis to experimentation.

awesome-autoresearch is a curated list of automated research use cases. This update adds two entries: Cribl's production deployment and Colab TPU port.

Jim Fan announced that the team launched ENPIRE, bringing AutoResearch into the physical world for the first time, equipping 8 Codex agents with robots, GPUs, and tokens, enabling them to autonomously learn and collaborate on real hardware to complete tasks.

awesome-autoresearch is a curated list of autoresearch use cases. This new entry applies Karpathy's autoresearch pattern to everyday software engineering scenarios, providing concrete templates to lower the adoption barrier for engineers.

Researchers benchmarked 7 frontier models on autoresearch tasks. Fable-5 won overall, but the open model Kimi-K2.7-Code surpassed others on ML engineering tasks.

Evo ported its autoresearch loop onto Anthropic's dynamic workflows in Claude Code, moving orchestration from in-context memory to deterministic JavaScript, solving long-horizon instruction adherence and enabling self-evolving workflows.

PostHog used an AI agent based on Karpathy's autoresearch to find a three-year-old bug in their ClickHouse query engine that prevented proper primary key usage for timestamp filters. Fixing it improved performance by 11% and reduced scanned granules by 62%.

yibie shared three new entries from the awesome-autoresearch list, covering automated quantitative trading, universal skill optimization, and a Claude Code plugin.

This week awesome-autoresearch added three items, including the autoslam project that applies Karpathy's autoresearch loop to LiDAR SLAM, and two blog posts analyzing the original experiments and revealing metric gaming.

This paper presents a two-level autoresearch framework where an outer-loop AI agent autonomously optimizes inner-loop LLM policy-synthesis pipelines for multi-agent sequential social dilemmas, achieving superior performance and discovering objective-specific mechanisms like fairness under a maximin welfare objective.

OpenAI is co-hosting an autoresearch hackathon this Saturday with Raindrop AI and Modal, focusing on building self-improving systems like agents and models.

This week, 9 new records were added to the autoresearch ecosystem, bringing the total to 383, covering multiple open-source tools and projects such as the AutoResearch-RL reinforcement learning framework, lance-autoresearch database kernel optimization, and Clio prediction market backtesting framework.

awesome-autoresearch updated, added dreamworld (world model research), Odyssey Engine (general iterative engine), and an article by Kirill Krainov on self-improvement of agentic coding.

Introduces the autoresearch project, which breaks down the AI research process into a verifiable loop (fixed environment, single editable file, fixed metric, Git rollback), enabling AI agents to perform controllable and reproducible experiment iterations; also mentions the 12-factor-agents checklist.