Copiloting

Polyglot AI/LLM experimentation monorepo with:

• TypeScript LangChain tutorials in tutorials/
• Python LangChain demos in course/sections/
• A Flask + SvelteKit PDF chat app in course/pdf-dist/

Prerequisites

• mise for local tool versions

mise.toml installs and pins the repo's Python, Poetry, pnpm, and Node versions.

Local Setup

Open the repo root in VS Code. The default Python workflow is repo-root based.

# Install pinned tool versions from mise.toml
mise install

# Create/use the repo Python 3.11 environment
poetry env use 3.11

# Install Python dependencies for the root project plus editable path deps
poetry install

# Install JS workspace dependencies
pnpm install --frozen-lockfile

VS Code

• Open the repository root, not a nested Python directory
• Install the recommended extensions (ms-python.python, ms-python.vscode-pylance)
• VS Code is configured to use the repo .venv
• Pylance is configured to resolve course/pdf-dist imports like from app.web import ...

If VS Code does not pick up the interpreter automatically, select .venv from the Python interpreter picker.

Python Commands

# Root Poetry scripts
poetry run agents
poetry run course
poetry run facts
poetry run facts-create-embeddings
poetry run tchat

# Flask DB init for the PDF app
poetry run flask --app app.web init-db

JavaScript Commands

pnpm build
pnpm check-format
pnpm -C tutorials run start:quickstart-llms
pnpm -C course/pdf-dist/client run dev

Notes

• Root Poetry setup covers the main Python work in course/sections and course/pdf-dist
• poetry.toml pins Poetry to an in-project virtualenv at ./.venv
• course/local-do is a separate optional Poetry project with its own setup
• This repo pins Python 3.11.7; your system python3 version does not matter if mise and Poetry are used correctly