Bart.dIAs - WebGRIPP AI Assistant

Bart.dIAs is an assistant for the WebGRIPP environment. Currently, it is solely an assistant for parallel programming that analyzes sequential code to identify bottlenecks and suggest parallelization strategies. It combines critical path analysis with pattern recognition to provide targeted recommendations for improving application performance.

Features

• ⚡ Bottleneck Analysis: Identifies sequential bottlenecks in Python code.
• 🔄 Parallelization Strategies: Suggests strategies to parallelize code using recognized patterns.
• 🎯 Critical Path Analysis: Combines CPA with pattern recognition for targeted recommendations.
• 🤖 LLM Chat: Conversational interface to discuss code optimizations.
• 📎 Code Context: Analyze open files directly from JupyterLab.

Prerequisites

• JupyterLab 4.x
• Node.js 18+
• Ollama running locally (default: http://localhost:11434)
• Bart.dIAs MCP Server (for deep code analysis)

Installation

1. Install Dependencies

cd bart_dias_jupyter_extension
jlpm install

2. Build Extension

jlpm build

3. Install in JupyterLab

jupyter labextension install .

Usage

1. Open the Bart.dIAs extension from the right sidebar.
2. Open a Python file you want to analyze.
3. Click the attach button (📎) to send the code to the assistant.
4. Ask for parallelization suggestions or performance analysis.

Project Structure

bart_dias_jupyter_extension/
├── src/                    # Frontend (React/TypeScript)
│   ├── index.ts            # Extension entry point
│   ├── components/         # UI Components
│   ├── services/           # API & MCP logic
│   └── hooks/              # State management
├── style/                  # CSS Styles
└── pyproject.toml          # Python Packaging

Technology Stack

• Frontend: React 18, TypeScript, Lumino
• Analysis: LangChain, MCP Adapters
• AI: Ollama (Local LLMs)

License

MIT