ScholarDevClaw reads ML research, maps improvements to your codebase, and generates validated patches — autonomously. Runs entirely on your machine.
From raw research paper to production pull request — entirely autonomous, with approval gates where it matters.
Tree-sitter extracts a full AST from your codebase across 6+ languages. Detects frameworks, patterns, and architecture.
tree-sitterSearches arXiv (2.4M+ papers), Papers with Code, and GitHub for ML research relevant to your specific codebase.
arXiv · PWC6-tier matching: exact name → fuzzy → imports → text scan → legacy → LLM semantic. Every match is confidence-scored.
6-tier15 CST templates (RMSNorm, FlashAttention, RoPE…) plus LLM synthesis fallback for novel improvements.
15 templatesReal subprocess benchmarks measure actual speedup and memory impact. Approval gates on low-confidence changes.
1260+ testsClean pull request on a safe branch with full changelog, benchmark results, and confidence scores attached.
git · CIProduction-grade tooling for teams who operationalize ML research at scale.
# 1. Install pip install scholardevclaw # 2. Point at your project scholardevclaw analyze ./my-transformer # → Found 6 improvement candidates # → RMSNorm: 94% confidence # → FlashAttn: 97% confidence # 3. Map paper spec → code scholardevclaw map . rmsnorm # 4. Generate + validate patch scholardevclaw generate . rmsnorm # → Generating patch artifacts... # → Running 1260 tests... PASS # → Speedup: +18.4% # 5. Or just launch the TUI scholardevclaw tui
Four install methods. Pick whichever fits your environment.
scholardevclaw demoscholardevclaw tuischolardevclaw analyze .Install ScholarDevClaw and run your first research-to-code pipeline in under a minute.