# WecoAI/awesome-autoresearch

Curated list of AutoResearch use cases with optimization traces and open source implementations

Repository: https://github.com/WecoAI/awesome-autoresearch
Canonical: https://ross.abutalabs.com/products/wecoai-awesome-autoresearch
License: CC0-1.0
License Family: permissive
Topics: ai-agents, ai-research, automated-machine-learning, autonomous-agents, autoresearch, awesome, awesome-list, claude-code, code-optimization, curated-list, llm, self-improving-ai
Last push: 2026-07-30T21:24:08+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 95, release rhythm 35, longevity 11
- inputs: {"age_days": 164, "days_push": 34, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1038, forks 77 (observed 2026-08-28T04:03:19.753345+00:00)

## What it is
A curated awesome-list of AutoResearch use cases, where a coding agent iteratively optimizes a file against an evaluation metric in a keep/discard loop. Each entry links to the actual optimization trajectory and open-source implementation, covering LLM training, GPU kernels, template engines, and prompt engineering.

## Use cases
- find examples of AI agents that automatically optimize code
- learn how the autoresearch optimization loop works
- see real optimization traces from LLM training experiments
- apply automated research loops to GPU kernel optimization
- find open-source implementations of self-improving AI workflows
- optimize prompts automatically with evaluation metrics
- discover community adaptations of Karpathy's autoresearch

## When to choose
- you want curated, trace-backed examples of agentic code optimization
- you're exploring how to adapt the autoresearch loop to new domains
- you need reference implementations before building your own automated research agent

## When to avoid
- you need a runnable tool or framework rather than a list of links
- you want guaranteed-maintained production software rather than community experiments
- you're not working with coding agents or GPU-based ML workflows

## Facets
- artifact type: learning-resource
- maturity: active
- function: agent-framework, llm-training, prompt-engineering, benchmarking, developer-tools
- domain: artificial-intelligence, machine-learning, awesome-lists, developer-tools, tutorials
- platform: python, cli, cross-platform
- tags: awesome-list, autoresearch, curated-list, optimization-traces, coding-agents, self-improving-ai, automated-ml, ai-agents, gpu

## Member repositories
- WecoAI/awesome-autoresearch (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:19.753345+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:04:14.018618+00:00, confidence not recorded.
  - readme: https://github.com/WecoAI/awesome-autoresearch (fetched 2026-08-28T04:03:19.753345+00:00, sha e2cca2325c9f)
- Data as of 2026-08-30T08:39:29.467469+00:00.