Hey Product Hunt community! 👋 As a software engineer who has worked in both big tech and on side projects, I've found that while AI coding assistants like Copilot and Cursor can boost development speed for personal projects, they often fall short when it comes to understanding and navigating the complexities of enterprise-level codebases. That's why I developed KnowLang (https://github.com/kimgb415/know...) - an open-source tool designed to help engineers gain a comprehensive understanding of large, intricate codebases and inter-repository dependencies. 🔍 How it works: 1. Parses and summarizes code 2. Indexes both code & summaries into a vector database 3. Provides a RAG (Retrieval-Augmented Generation) chatbot interface for easy interaction and exploration 💡 Key features and plans: 1. Code-specific RAG systems with code-focused reranking and embeddings 2. Automatic LLM fine-tuning to reduce sole reliance on RAG 3. Inter-repository knowledge awareness through multi-layer RAG 🙌 Try KnowLang: 1. Explore on GitHub: https://github.com/kimgb415/know... 2. Install the package: pip install knowlang 3. Demo on Hugging Face Space: https://huggingface.co/spaces/ga... 📝 Learn more: Read the detailed blog post about KnowLang's architecture and performance on the Hugging Face blog: https://huggingface.co/blog/gaby... I'm excited to share KnowLang with the Product Hunt community and would love to hear your thoughts, feedback, and suggestions. Let's work together to make code comprehension more accessible and efficient for developers everywhere! 🚀 Looking forward to your insights!

KnowLang

Open-source tool to make complex codebases understandable.

Open-source tool to make complex codebases understandable.