burnt

Burnt’s building an agentic OS for the Food Supply Chain with AI Agents for each department. Our first AI Agent replaces the whole order management team at food distribution companies that sell into restaurants.

Our agent monitors all incoming messages, figures out if it's an order, learns each customer's habits, shorthand, and favorite products, picks up on any prep notes, and enters it straight into the existing ERP. The more it sees, the smarter it gets.

At Burnt, we’re building an Agentic operating system for the food supply chain, replacing slow, error-prone ERP workflows with intelligent agents across sales, operations, credit control, and implementation. Distributors today wrestle with outdated systems, manual order entry from emails and WhatsApp, credit teams buried in follow-ups, and inventory locked up due to disconnected processes—resulting in lost revenue and rising operational costs. Even worse, traditional ERP implementations take 12–18 months and often fail. We're building implementation agents that reduce go-live time to weeks by auto-mapping SKUs, learning from customer data, and adapting to edge cases through feedback loops. Technically, this means tackling messy, unstructured inputs, real-time decisioning, multi-tenant architecture, and long-term memory across agents. Our stack includes Java (Spring Boot), Python, AWS , Pinecone for semantic search, and LLMs from OpenAI and Bedrock. If you’re excited by vertical SaaS, applied AI, and building intelligent systems that do real work, we’d love to talk.