Deep Diver Setup: Native
For experienced developers who want full control and the ability to browse, run, debug, and tweak the code.
Prerequisites
- Development experience with basic command line and version control
- Python 3.12.x installed and accessible via
python
command - Git installed
- Your preferred development environment (IDE, editor, shell, etc.)
Setup
1. Clone and navigate to the repository
2. Verify Python version
Should show Python 3.12.x
. If you only have python3
, feel free to use
that throughout.
3. Set up Python dependencies
Choose your preferred approach:
Option A: Using Poetry (if you have it)
Option B: Using venv
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
python -m pip install -e .
4. Prepare environment variables
5. Verify setup
This should print a welcome message and confirm all dependencies are working.
6. Optional: Set up your own LangSmith account
If you want to debug your own traces instead of using shared workshop access:
- Sign up for a free account at https://smith.langchain.com
- Get your API key from the account settings
- Add
LANGCHAIN_API_KEY
to your.env
file
During the Workshop
1. Get latest code
2. Activate your environment (if not using an IDE that handles this)
Poetry: poetry shell
venv: source .venv/bin/activate
3. Update dependencies
poetry install
or python -m pip install -e .
4. Add API keys to .env
file (provided during workshop)
5. Run the application
This opens the Chainlit interface in your browser.
Debugging and Observability
You'll participate in debugging exercises using LangSmith:
- Shared access: We'll provide credentials to view workshop traces
- Your own traces: If you set up your own account above, you'll see your own traces
LangSmith access: https://smith.langchain.com
Troubleshooting
- Python version issues: We need 3.12.x specifically for compatibility with all dependencies
- Virtual environment not active: Make sure your terminal prompt shows
(.venv)
or similar - Permission errors with
ws
command: Trypoetry run ws hello
instead (if using Poetry) - Import errors: Run
python -m pip install -e .
again
After the Workshop
Cleanup
If you no longer need this environment:
This removes everything - the code, virtual environment, and all dependencies.
Ready to dive deep! 🤿