Well

une document understanding models Example use case: You can ask the model to generate a receipt from a Vietnamese coffee shop, with broken math and a typo in the merchant name — and get structured JSON output that mimics what real OCR might return. Would love feedback from folks working on: LLM eval tooling Synthetic data generation OCR / document AI AI agents that touch financial data Happy to support anyone who wants to extend it or wire up other backends (Claude, Mistral, LM Studio, etc).

ast, standardized, and predictable, handling supplier invoices remained chaos. --- The Insight Payments today are a solved problem. Protocols like EMV, SEPA Instant, and SWIFT create reliable, instant settlement flows. Whether you tap a card or wire funds across borders, the infrastructure just works. But invoices? Every supplier, every service, every vendor still does it differently. Some email you a PDF. Some force you to log into portals. Some send nothing at all unless you chase them manually. There’s no protocol, no standards, no rails — just friction. In a world where payment is protocolized, *invoice management is broken*. --- The Problem This chaos isn’t just annoying — it’s operationally expensive: - Solopreneurs and small teams waste hours every month just retrieving invoices. - Accounting tools often rely on manual drag-and-drop uploads. - Poor invoice tracking means poor treasury visibility — contributing to the 57% of bankruptcies in France caused by cash flow mismanagement. We’ve seen it first-hand working closely with thousands of startups and SMEs at Qonto. Adoption of accounts payable solutions is painfully low — not because people don’t need them, but because getting started still requires too much manual work. No one wants to forward emails, upload PDFs, or manage inbox rules just to track basic finances. --- Why Now Two forces are converging: - Regulation: EU-wide electronic invoicing mandates are coming, shifting the landscape. - Technology: AI now allows us to automate where legacy RPA (robotic process automation) approaches failed. We believe the next true disruption in finance won’t come from payment rails. It will come from *bridging the invoice gap* — automatically, invisibly, and at scale. --- What We’re Building Well is an AI-powered infrastructure designed to automate supplier invoice collection — no matter the format, no matter the source. Here’s what Well does: • Captures invoices from portals, emails, and attachments — fully automated. • Extracts structured data (amounts, vendors, dates) with high accuracy. • Securely syncs invoice data into your accounting SaaS, ERP, or dashboards. No manual uploads. No password sharing. No chasing. Just a protocolized pipeline — like payments, but for supplier invoices. --- What’s Next - Extending Well’s reach across 1,000+ SaaS vendors. - Deepening integrations into accounting, ERP, and spend management ecosystems. - Building invisible finance workflows where invoices just...appear where you need them. We’re starting with solopreneurs, indie hackers, and lean startup teams — but our ambition is to make invoice chaos obsolete for businesses of any size. If you’re tired of chasing supplier invoices instead of growing your business, we’d love to hear from you. Learn more at https://wellapp.ai/