Role Overview
We’re seeking a full-stack developer with demonstrated experience shipping production AI products powered by large language models or other foundation models. The project focuses on extracting structured data from documents and making it usable by enterprise systems, including SAP and other ERPs, through structured APIs and data models. This is not a chatbot or Q&A-focused role.
Responsibilities
- Build and ship production software that uses foundation models to extract structured data from documents.
- Connect extracted data to enterprise systems, including SAP and other ERPs, using their structured APIs and data models.
- Develop and maintain production software using AWS serverless, Angular, and Node.js.
- Write automated unit tests for predictable, rules-based components and automated evaluations (evals) for AI components with variable outputs.
- Build MCP servers from scratch or, where relevant, develop on AWS Bedrock AgentCore MCP Gateway using Cognito, DynamoDB, and Lambdas.
Required Skills
- Proven experience building and shipping production software powered by large language models or other foundation models.
- Experience extracting structured data from documents for use in enterprise systems; chatbot or Q&A experience alone is not sufficient.
- Production experience with AWS serverless, Angular, and Node.js, demonstrated through shipped projects.
- Experience writing automated unit tests for rules-based functionality and evals for AI functionality.
- Experience using Claude Code or similar tools to write software. Tool use alone does not replace production AI product experience.
- Experience building MCP servers from scratch.
- Experience with AWS Bedrock AgentCore MCP Gateway, including Cognito, DynamoDB, and Lambdas.
Nice to Have
- Experience connecting document-extraction outputs to SAP or other ERP systems.
- Experience with both custom MCP server development and AWS Bedrock AgentCore MCP Gateway.