**Hey Product Hunt! 👋 I'm Anirudh, the maker of FeynmanWiki.** Reading machine learning papers can be dense. Whether diving into new reinforcement learning architectures or scaling laws, building a true "first-principles" intuition often takes hours of deciphering walls of text and equations without enough visual aids. I built **FeynmanWiki** to solve this. It utilizes a multi-agent AI architecture to transform any ML paper, concept, or topic into a comprehensive, diagram-rich deep dive. **Here is what you can do with it:** * **Generate Visual Deep Dives:** Turn a prompt (like "Diffusion and flow matching") or a PDF upload into a beautifully structured article complete with custom diagrams and accurate math. * **Multi-Agent Orchestration:** The generation process is split into specialized tasks (curriculum planning, writing, topic checking, and image generation). * **Bring Your Own Key (BYOK) & Model:** You have complete control under the hood. Securely plug in your API keys for OpenAI, Anthropic, or DeepSeek. * **Granular Cost & Quality Control:** Want Claude 3.5 Sonnet for writing but DeepSeek V4 for the curriculum outline? You can mix and match models per agent to optimize for your budget and desired quality, with real-time cost estimations per slide. * **Curated Library & Workspaces:** Organize your generated deep dives and explore a community library of trending ML topics wrapped in a clean, minimalist UI. I’d love to hear your feedback on the generated diagrams, the multi-agent workflow, and the overall reading experience. What paper or topic are you going to run through it first? I'll be hanging out in the comments all day to answer any questions! 🚀

FeynmanWiki

Generate diagram-rich ML blogs from any paper or topic

Generate diagram-rich ML blogs from any paper or topic