Senior Software Engineer

Burnt is building the AI operating brain for the supply chain. Food distribution is our entry point: a trillion-dollar industry that still runs on decades-old ERPs, spreadsheets, and manual work at every layer of the operation.

We don't sell software that assists people with that work. We deploy AI agents that own entire job functions. Our customer service agent runs the CS desk, reading inbound orders across email, text, phone, and fax and writing them directly into the customer's ERP, handling the large majority of inbound volume autonomously and escalating to a human only when judgment is required.

Our procurement agent runs the buying desk: demand forecasting, inventory and vendor lead-time tracking, and building each morning's purchase orders for buyer approval. Beneath the agents sits a world model assembled from each customer's operational systems and communications, and every new agent we ship inherits it.

The roadmap is deliberate: an agent in every department, and ultimately customers running their entire operation on Burnt. We were founded by operators with generations of family history in the food supply chain and are backed by Penny Jar Capital, Scribble Ventures, and Formation VC. Our customers today range from regional distributors to food businesses with billions in revenue.

About Burnt

Burnt is building the AI operating brain for the supply chain. Food distribution is our entry point: a trillion-dollar industry that still runs on decades-old ERPs, spreadsheets, and manual work at every layer of the operation.

We don't sell software that assists people with that work. We deploy AI agents that own entire job functions. Our customer service agent runs the CS desk, reading inbound orders across email, text, phone, and fax and writing them directly into the customer's ERP, handling the large majority of inbound volume autonomously and escalating to a human only when judgment is required.

Our procurement agent runs the buying desk: demand forecasting, inventory and vendor lead-time tracking, and building each morning's purchase orders for buyer approval. Beneath the agents sits a world model assembled from each customer's operational systems and communications, and every new agent we ship inherits it.

The roadmap is deliberate: an agent in every department, and ultimately customers running their entire operation on Burnt. We were founded by operators with generations of family history in the food supply chain and are backed by Penny Jar Capital, Scribble Ventures, and Formation VC. Our customers today range from regional distributors to food businesses with billions in revenue.

• Production AI agents that run operational workflows end to end for food distributors
• Full-stack features in Type Script, NestJS, and React, shipped to production and maintained there
• Eval frameworks and feedback loops that measure and improve agent performance over time
• Large-scale order, catalog, and operational data at the application layer
• AWS infrastructure (Lambda, ECS, RDS, S3) managed with Terraform
• Observability as a first-class concern, with logging, tracing, and alerting built in from the start

We are looking for 5 to 7 years of experience and all of the following:

• Node.js, Type Script, NestJS, and React in production (required)
• At least one AI agent built, deployed, and running in production, beyond the prototype stage (required)
• Applications deployed on AWS and serving real traffic (required)
• Hands-on experience with LLM and agent frameworks such as Lang Chain or Llama Index
• Production use of observability tooling such as Datadog, Open Telemetry, or Cloud Watch
• Experience designing eval frameworks for AI systems and building self-learning or feedback-loop systems
• Direct experience managing production incidents end to end, from detection through resolution
• The ability to clearly articulate and defend system design decisions you personally made
• Active use of AI coding tools such as Cursor, Claude Code, or Copilot to increase output, grounded in a track record of writing production code before LLM-assisted development existed

• Intro call with the CEO and CTO
• Technical interview 1
• Technical interview 2
• Culture interview
• Offer

• The supply chain is one of the largest industries in the world, and the systems layer underneath it has not been rebuilt in decades. AI agents capable of running these workflows end to end have only recently become possible, and the team that executes on this now will define the category. We intend to be that team, and this role is central to it.