DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

How to share a self-serve data explorer as a Streamlit app with Pygwalker

Host the PyGWalker UI in Streamlit so teammates can explore a dataset in the browser without a notebook.

Short answer: Install Pygwalker and streamlit; then import StreamlitRenderer from Pygwalker.api.streamlit and set the page layout to wide; Wrap the renderer in a function decorated with st.cache_resource so memory does not explode. All steps.

Who it is forAn analyst whose colleagues keep asking for slightly different cuts of the same data.

The problemEvery new question becomes a request to the one person who has the notebook.

The outcomeA web app where anyone can drag fields to build their own charts on the shared dataset.

What you need first

  • Python with pandas, polars or pyarrow data
  • Jupyter Notebook, JupyterLab, Colab or Kaggle, or Streamlit for a web app
  • pip, conda or mamba to install

Steps

  1. Install Pygwalker and streamlit
  2. Import StreamlitRenderer from Pygwalker.api.streamlit and set the page layout to wide
  3. Wrap the renderer in a function decorated with st.cache_resource so memory does not explode
  4. Create StreamlitRenderer(df, spec_path="./gw_config.json", spec_io_mode="rw") to let users save chart configs
  5. Call renderer.explorer() and run the app with Streamlit

Prompt for your agent

Build a Streamlit app with Kanaries/pygwalker that loads <dataset>, caches the StreamlitRenderer with st.cache_resource, saves chart configs to gw_config.json, and shows the explorer. Use computation="kernel" if the data is large.

Caveat. With spec_io_mode set to rw, anyone using the app can change the saved chart config, so keep a backup.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for Kanaries/pygwalker and "Share a self-serve data explorer as a Streamlit app"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer