Radical Whale
Build and enrich datasets with your LLM and MCP integrations
8 followers
Build and enrich datasets with your LLM and MCP integrations
8 followers
Build, enrich, and automate datasets using your LLM keys, APIs, MCP servers, and agents. No markup, no seats—just intelligent data workflows.
Hey everyone 👋 Nick here — creator of Radical Whale.
Radical Whale is an AI-native data workspace that lets you build, enrich, and automate datasets using your own LLM keys, your own APIs, and your own agents.
You can define datasets with AI columns that determine values based on instructions, build custom AI agents that can research and enrich data, connect any API or MCP tool like Tavily, Github, Apollo, Diffbot 🌐, collaborate in TipTap-powered notebooks that mix text, datasets, and agent calls 🧠, and run everything in dedicated queues per workspace — fast, isolated, and predictable ⚡️
Why I built this 💡
I was frustrated with platforms like Attio, Notion, and Freckle that abstract away the real agents, hide the true cost of LLM calls, and sell you “credits” on top. I knew I could run similar workflows for pennies by calling the models directly and using the APIs I already pay for.
Radical Whale is the platform I wanted — a way to build and scale intelligent data workflows without middlemen or markup 🚀
If you work with structured data, enrichment, or AI automation, I’d love for you to try it and share your thoughts 🙏
https://radicalwhale.com — let’s make AI infrastructure simple, open, and affordable 🌊
Where is the data housed? A vector DB?
@thareja We use postgres to store the data. Nothing is vectorized at the moment, although we have plans on the roadmap to add a knowledge base feature to Radical Whale.