Erdos is an open-source, AI-powered IDE for data scientists, forked from VS Code. It integrates AI assistance for Python, R, SQL, and Julia across scripts and Jupyter notebooks, along with dedicated panes for plots, databases, environments, and help documentation. The makers claim existing AI coding tools neglect data scientists, creating a need for a specialized IDE.
A starter prompt for Claude Code, what you'll need, and how to reach them.
You are an expert software engineer. I need to understand the feasibility of creating a specialized AI-powered IDE for data scientists, forked from VS Code. This IDE, provisionally named 'DataForge', should integrate AI assistance across Python, R, SQL, and Julia within scripts and Jupyter notebooks, and include specialized panes for plots, databases, environments, and help documentation. Our core stack is Next.js 16 App Router, React 19, Tailwind v4, AI SDK v6 with Gemini, and Neon Postgres on Vercel, but for an IDE, we'll need to work within the VS Code extension architecture and potentially Electron. Your task is to outline the most critical technical challenges and architectural decisions for this fork, focusing on the AI integration layer. Specifically, detail how to approach: 1. **AI Integration for Multi-Language/Multi-File Type:** How would the AI effectively 'search, read, and write across all file types' (Python, R, SQL, Julia, Jupyter notebooks)? Focus on the 'jupytext system' optimization for notebooks. 2. **Specialized Panes Implementation:** How to integrate Plot, Database, Environment, and Help panes within the VS Code fork architecture, ensuring they are interactive and context-aware. 3. **Local/BYO Key AI Backend:** Design a secure, zero-data-retention mechanism for AI model inference, supporting both local models and external APIs (e.g., Anthropic, OpenAI) via a 'bring your own key' approach. 4. **Remote Development:** Key considerations for implementing SSH and container-based remote development. For the MVP, focus on Python/Jupyter notebook support with AI code generation/refactoring and an integrated plot pane. Provide a detailed, step-by-step technical plan for these initial features. Conclude with a 'build/verify' gate: how would I confirm these core features are working as expected?
Hey HN! We’re Jorge and Will from Lotas ( https://www.lotas.ai/ ), and we’ve built Erdos, a secure AI-powered data science IDE that’s fully open source ( https://www.lotas.ai/erdos ). A few months ago, we shared Rao, an AI coding assistant for RStudio ( https://news.ycombinator.com/item?id=44638510 ). We built Rao to bring the Cursor-like experience to RStudio users. Now we want to take the next step and deliver a tool for the entire data science community that handles Python, R, SQL, and Julia workflows. Erdos is a fork of VS Code designed for data science. It includes: - An AI that can search, read, and write across all file types for Python, R, SQL, and Julia. Also, for Jupyter notebooks, we’ve optimized a jupytext system to allow the AI to make faster edits. - Built-in Python, R, and Julia consoles accessible to both the user and AI - Plot pane that tracks and organizes plots by file and time - Database pane for connecting to and manipulating SQL or FTP data sources - Environment pane for viewing variables, packages, and environments - Help pane for Python, R, and Julia documentation - Remote development via SSH or containers - AI as
Reply in the Hacker News thread, and consider also reaching out via email to jorgeoguerra@lotas.ai, as listed on their site.
“I'm a solo founder with strong AI development skills, and I'm deeply impressed by Erdos. I've built a prototype demonstrating enhanced AI interaction for Jupyter notebooks within a VS Code-like environment. I'd love to contribute to Erdos, especially on the AI integration or specialized pane development, and explore how I might support its growth.”
Open the original ↗