Add a new Real-World Projects lesson: Codebase Q&A — "Where Is X Implemented?" (backlog idea #18).
Scope
docs/projects/codebase-qa/index.md — new lesson matching the house skeleton. This is the practical sibling to the existing codebase-knowledge-graph project (AST + graph) and the RAG-over-GitHub-repo project (embeddings + citations). Its focus: answer "where is X implemented / how does X work" questions over a real repo using a lightweight two-stage search — symbol/AST-level matching for precise "where is this defined" questions, plus embedding search for fuzzy "how does this concept work" questions — then ground the answer with an LLM that must cite file:line.examples/codebase-qa/ — a real, runnable tool: clones (or points at) a local repo, builds a small AST-level index of symbols (function/class definitions, imports) with the built-in ast module, and a separate embedding index of code + docs, then answers two kinds of questions: exact ("where is X defined?") and semantic ("how does X work?"). Includes a bundled sample_repo/ for out-of-the-box testing, a notebook.ipynb, and a README.md.- Registration:
src/data/projects.ts, docs/projects/index.mdx, src/pages/index.tsx (English only — no i18n changes).
- Colab/Kaggle/Binder badges in "Where to run this", pointed at this PR's branch (to be updated to
main once merged); the notebook writes a tiny sample repo inline (hosted notebooks have no local files) and runs both query modes over it.
Framing note
The lesson should be honest that "where is X implemented?" is really two different questions with different best tools — exact-symbol lookup (where definitions/imports live) vs. semantic search (what code is about) — and that a good tool offers both and says which it's answering. It should encourage verifying answers against the actual cited lines, and acknowledge that symbol search is language-specific (this project does Python's AST; other languages need their own parser).
Add a new Real-World Projects lesson: Codebase Q&A — "Where Is X Implemented?" (backlog idea #18).
Scope
docs/projects/codebase-qa/index.md— new lesson matching the house skeleton. This is the practical sibling to the existing codebase-knowledge-graph project (AST + graph) and the RAG-over-GitHub-repo project (embeddings + citations). Its focus: answer "where is X implemented / how does X work" questions over a real repo using a lightweight two-stage search — symbol/AST-level matching for precise "where is this defined" questions, plus embedding search for fuzzy "how does this concept work" questions — then ground the answer with an LLM that must cite file:line.examples/codebase-qa/— a real, runnable tool: clones (or points at) a local repo, builds a small AST-level index of symbols (function/class definitions, imports) with the built-inastmodule, and a separate embedding index of code + docs, then answers two kinds of questions: exact ("where is X defined?") and semantic ("how does X work?"). Includes a bundledsample_repo/for out-of-the-box testing, anotebook.ipynb, and a README.md.src/data/projects.ts,docs/projects/index.mdx,src/pages/index.tsx(English only — no i18n changes).mainonce merged); the notebook writes a tiny sample repo inline (hosted notebooks have no local files) and runs both query modes over it.Framing note
The lesson should be honest that "where is X implemented?" is really two different questions with different best tools — exact-symbol lookup (where definitions/imports live) vs. semantic search (what code is about) — and that a good tool offers both and says which it's answering. It should encourage verifying answers against the actual cited lines, and acknowledge that symbol search is language-specific (this project does Python's AST; other languages need their own parser).