Enhancing your JupyterLab Developer Experience with Local LLMs and Code Snippets
summary
Demo shows a JupyterLab extension that captures code, generates runnable snippets, and answers errors using local models. It uses Pieces extension inside JupyterLab and switches between cloud GPT and on-device LLMs (Llama 2, Code Llama). Example fixes a broken sklearn import, auto‑generates a confusion‑matrix plot snippet, and saves a sharable snippet with description and links. Workflow runs locally, preserves privacy, and the demo shows it working end to end. save and share runnable snippets
AI-written summary · about
chapters (6)
- 0:00Enhancing JupyterLab with LLMs
- 1:09Pain points in data science
- 8:55Collaboration with code snippets
- 13:28Live demo: Pieces for JupyterLab
- 27:09Benefits and team setup
- 28:35Q&A
related talks
45mEverything You Need to Know About Running LLMs Locally
29mPodman AI Lab: For developers to build AI Applications with LLMs running locally 
37mEmpowering Developers with local AI
54mThe secret weapon for developers, Notebooks.
40mCreating Interactive GenAI applications powered by Streamlit, open source LLMs
01h 15m10x Development: LLMs For the working Programmer
39mAI Assistance Beyond Code: What Do We Need to Make it Work?
42mSimplifying Data Analysis with GitHub Codespaces, Jupyter Notebooks & Open AI