Key Features
Tech Stack
You can:
• Create datasets with AI columns that compute values from instructions
• Build custom agents that call any API or MCP tool (Tavily, GitHub, Apollo, Diffbot, etc.)
• Mix text, datasets, and agent calls in TipTap-powered notebooks
• Run workloads in isolated per-workspace queues for predictable performance
I built this because tools like Attio, Notion, and Freckle hide the real agent calls behind credit systems and markup. By calling models and APIs directly, you can operate these workflows far more efficiently and transparently.
If you work with structured data, enrichment, or AI automation, I’d appreciate feedback on the approach.
Not affiliated with Hacker News or Y Combinator. We simply enrich the public API with analytics.