Retab came out with new funding, new product, and a plan to become infrastructure for the next wave of vertical AI.
Retab, a developer-first platform for document automation, launched out of stealth with US$3.5 million in pre-seed funding and a bold mission to become a core infrastructure layer for the AI economy.
Designed to address the limitations of existing document AI systems, Retab offers a robust, production-ready solution for extracting structured data from unstructured documents at scale.
Backed by investors including VentureFriends, Kima Ventures, and notable tech figures, the funding will accelerate platform development and community growth as demand increases across vertical AI startups and enterprise automation teams.
Retab enables developers to define the data schema they need, with the platform automating the entire pipeline, from labelling and evaluations to prompt engineering and model selection.
Positioned as a middleware intelligence layer, Retab integrates with top-tier models from providers like OpenAI, Google, and Anthropic, transforming them into production-ready systems for document-heavy workflows.
The platform includes self-optimising schemas, where an AI agent refines instructions for maximum accuracy; intelligent model routing, which benchmarks and routes tasks to the most efficient model; and guided reasoning with k-LLM consensus, where step-by-step thinking and multiple-model validation improve reliability and reduce uncertainty. These features collectively help reduce costs by up to 100x over traditional document extraction systems.
“People keep building demos that look like magic, but break the moment you put them into production,” said Louis de Benoist, co-founder and CEO of Retab. “We lived that pain ourselves. Wiring up fragile pipelines just to extract a few fields from a PDF. We built Retab because it’s the developer-first platform we always wished we had.”
The founders first encountered this problem while building internal automation tools in the logistics sector. The orchestration layer they developed to manage unreliable models later became the foundation for Retab. Today, it is used by companies across logistics, finance, and healthcare to automate critical workflows, from claims processing and identity verification to extracting complex data from multi-page financial reports.
Customers have already reported measurable benefits. A major trucking firm, for example, identified the smallest model configuration that met their 99% accuracy target, significantly cutting costs. A financial services provider now uses Retab to parse 200-page quarterly reports, a task that once took teams of analysts days to complete.
Florian Douetteau, co-founder and CEO of Dataiku and investor in Retab, said: “The AI-fication of the economy depends on the capability to convert operations based on millions of documents into verified, structured data that autonomous systems can utilise. The team at Retab understands this thoroughly and is uniquely positioned to solve it for the thousands of AI first companies that are emerging.”
Looking ahead, Retab is expanding its capabilities beyond documents, with upcoming support for structured extraction from websites and integrations with tools like n8n, Zapier, and Dify.
