amoslee2026

amoslee2026 / Babel

Public

AI-native Chiplet design flow based on open-source EDA toolchain

20
3
89% credibility
Found May 23, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
SystemVerilog
AI Summary

Babel is an open-source AI-powered system that helps engineers design computer chips (specifically Chiplets) by coordinating multiple AI agents. Each AI agent specializes in a different stage of chip design—from understanding what you want to build, to creating the detailed chip architecture, writing the actual circuit code, verifying it works correctly, and finally producing the manufacturing files. The system includes quality checks at each stage and can automatically fix problems. It uses well-known open-source chip design tools and is designed to work with professional 7nm manufacturing technology. The project also includes training materials to help people learn how to build similar AI-powered automation systems.

How It Works

1
💡 Learn about AI chip design

You discover Babel through documentation or word of mouth, learning it uses AI to help design complex computer chips automatically.

2
🔧 Set up your workspace

You install the open-source tools and connect your AI assistant to the design environment, following simple setup instructions.

3
📝 Describe what you want to build

You tell the AI what kind of chip you need, like an AI processor that runs at a certain speed and handles specific tasks.

4
🤖 AI agents collaborate on your design

Multiple AI specialists work together: one designs the architecture, another writes the chip code, another verifies it works correctly, and more.

5
Design quality check
✅
Design passes quality check

Your design meets all requirements and moves forward to the next stage

🔧
Needs improvement

The AI identifies issues and automatically fixes them before continuing

📊 Get your chip layout ready

After all the work, you receive the final chip layout file (GDSII) that can be sent to a factory for manufacturing.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Babel?

Babel is an AI-native chip design workflow that automates the entire Chiplet development pipeline—from architecture specification through RTL generation, verification, synthesis, and physical design—using a multi-agent system powered by Claude Code. It integrates open-source EDA tools like Yosys, Verilator, OpenSTA, and Magic to deliver a complete, vendor-neutral design flow targeting the ASAP7 7nm process node. Users describe designs in plain language, and the system generates specifications, RTL code, testbenches, and timing constraints automatically.

Why is it gaining traction?

The chip design industry has long been dominated by expensive, closed-source tools with steep learning curves. Babel stands out by combining AI-driven automation with a fully open-source toolchain, making tape-out capabilities accessible to teams without commercial EDA budgets. Its 5-guru pipeline enforces design discipline through quality gates and handoff protocols, while supporting multiple clock and reset domains. The project also ships with comprehensive document templates for chiplet architecture and training labs for harness engineering—a rare combination that serves both production and education.

Who should use this?

Hardware engineers prototyping chiplets on a budget will benefit most. Research teams exploring open PDK flows without licensing overhead can evaluate tape-out可行性 without committing to commercial toolchains. Academic groups teaching chip architecture can use the training materials as curriculum. Startups prototyping early silicon will appreciate the rapid iteration enabled by the AI-driven workflow. The training labs are particularly valuable for engineers new to AI coding agents in EDA contexts.

Verdict

Babel shows impressive technical depth with its agent pipeline and skill taxonomy. However, with only 20 stars and a credibility score of 0.9%, this is clearly an early-stage project. The documentation and training materials are well-structured, which suggests active development, but production readiness remains unproven. Teams wanting to experiment with AI-driven chip design on open-source toolchains should absolutely give it a look—just do not bet your tape-out schedule on it yet.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.