Create a custom, terminal-based tool to automatically download, transcribe, clean, and summarize your favorite podcasts. This system extracts key insights, actionable quotes, and company mentions, turning audio content into a structured, readable format.
with Claire Vo
Set up a script that automatically downloads the latest episodes of your target podcasts. Use a tool like ffmpeg to handle the audio files and a local transcription model like Nvidia's Parakeet (run on a Mac) or OpenAI's Whisper for fast and accurate audio-to-text conversion.
The initial transcription will contain filler words and conversational tics. Use a local LLM like Gemma 3 (via Ollama) to clean up the text while preserving all important content and technical details.
You're a transcript editor. Clean up this podcast while preserving all the content. Keep the same length, remove the ums and the ahs, preserve all technical conversations.Create a main orchestrator script to manage the daily processing workflow. Store the cleaned transcripts in a local database like DuckDB to keep a persistent record and easily track which episodes have already been processed.
Feed the cleaned transcripts into a powerful LLM to generate a detailed, structured summary. The goal is to extract the most valuable information in a consistent format.
Prompt the LLM to structure its output with specific sections to ensure consistency and usefulness. Include fields like: Host and Guest, Comprehensive Summary, Key Topics and Themes, Actionable Quotes, Investment Theses, Noteworthy Observations, and Company Mentions.
Instead of relying on traditional Named Entity Recognition (NER) libraries, use the large language model to identify and list any startups or established companies mentioned in the podcast. LLMs often perform this task more accurately with less pre-processing.
Follow Sahil Lavingia's method for delegating small but impactful maintenance tasks, like fixing an internal Slack report, to the AI agent Devin. This workflow shows how to use simple, clear prompts to fix code and update system prompts without distracting your engineering team.
Learn Gumroad CEO Sahil Lavingia's process for rapidly prototyping a UI with v0, then using the AI agent Devin to autonomously code, implement, and submit the feature for review, turning a multi-day task into a one-hour workflow.
Improve your writing in real-time by using ChatGPT as a first reader. Get instant feedback on structure, clarity, and flow as you write, allowing you to polish your work section by section.
We've made ChatPRD affordable so everyone from engineers to founders to Chief Product Officers can benefit from an AI PM.