San Francisco-based AI startup Retab has raised $3.5 million in pre-seed funding to accelerate its mission of making document AI production-ready. The round was led by VentureFriends, Kima Ventures, and K5 Global, with participation from prominent angel investors including Eric Schmidt (via StemAI), Olivier Pomel (CEO, Datadog), and Florian Douetteau (CEO, Dataiku).
Founded by automation engineers Louis de Benoist and team, Retab provides a developer-first platform that extracts structured data from complex documents. Rather than offering its own LLM, Retab acts as an orchestration layer, optimizing prompt engineering, model routing, and error handling across providers like OpenAI, Google, and Anthropic. Already adopted by dozens of companies, Retab ensures scalable, verifiable document processing for critical workflows. The company is now expanding into website data extraction and launching integrations with Zapier, n8n, and Dify.
Featured image: Credit: Retab
