Lenny RAG

About

The Lenny RAG server offers a powerful hierarchical RAG system across 299 transcripts from the Lenny Rachitsky podcast. It's specifically designed to accelerate product development brainstorming by allowing users to semantically search for relevant insights, real-world examples, and access full episode context efficiently. The system structures podcast content into a multi-level hierarchy—episodes, topics, insights, and examples—which optimizes context window usage for large language models and facilitates progressive disclosure, enabling deep dives into specific areas only when needed.

Key Features

• Full transcript loading with episode metadata and topic breakdown.
• 3 GitHub stars
• Semantic search across 299 Lenny Rachitsky podcast transcripts.
• Detailed retrieval of specific topic context including summaries, insights, examples, and raw transcript segments.
• Hierarchical data extraction for efficient context management (Episode, Topic, Insight, Example).
• Implicit anchor detection to infer unnamed companies and examples based on guest background.

Use Cases

• Deep diving into specific topics or episodes from the Lenny Rachitsky podcast.
• Finding actionable insights and lessons from industry leaders presented in the podcasts.
• Product development brainstorming to retrieve relevant insights, real-world examples, and full transcript context.