Data-Sculptor

An integrated IDE agent that mentors users in Machine Learning and Data Science by providing real-time educational feedback

Features

1. Real-time syntax analysis

   Get instant feedback on your code quality with advanced linters.

2. Deep Syntax Validation

   Comprehensive code analysis using multiple linters: pylint, mypy, dodgy, pydocstyle, vulture

3. Integrated Workflow

   End-to-end pipeline from code editing to validation:

   • Edit code in Jupyter notebooks

   • Press Syntactic-Analysis button and get the feedback in LSP format

   • Press Semantic-Analysis button and get non-localized feedback as the .md report in same directory with .ipynb and localized feedback as comments inside the code cell

4. One-Click Environment

   Fully containerized setup with Docker Compose

🚀 How to use

1. Clone the repository:

git clone https://github.com/IU-Capstone-Project-2025/Data-Sculptor.git
cd Data-Sculptor/deployment/uat

2. Start the JupyterHub

docker compose -p uat --env-file uat.env up --build -d

3. Enter the JupyterHub container

docker exec -it uat-jupyterhub bash

Then:

mkdir -p /home/developer/.local/share/jupyter/runtime && \
chown -R developer:developer /home/developer

❗️Exit the container before authorization in browser

4. Login using authorization data:
   • http://localhost:11000

⚙️ Requirements

• docker version ~28.0.4, can be installed from the official Docker website
• if the user from which you are trying to run containers is not in Docker group - you should use \

sudo docker compose -p uat --env-file uat.env up --build -d