AI / Agent Engineer

Role overview Build and operate AI agents that solve practical problems in production. You’ll take systems from early scoping through deployment and ongoing support, combining agent workflows and data retrieval with the reliability measures needed for real-world use.

Responsibilities - Design agent workflows that use memory, tools, and human review, using frameworks such as LangGraph or Strands and cloud agent services. - Connect agents to relational, vector, and graph data so responses can be relevant, current, and traceable to their sources. - Develop evaluations and observability, along with retries, safeguards, and least-privilege access controls. - Work with clients and teammates to scope, ship, measure, and support solutions.

Requirements - Strong TypeScript/Node.js and/or Python skills, with experience taking LLM or agent systems into production. - Practical judgment about when AI is useful and when conventional software is a better fit. - Experience with AWS, CI/CD, and production concerns such as cost, latency, security, and failure handling. - A measurable approach to system quality, including interest in evaluations and observability.

Nice to have - Experience building with the Model Context Protocol or working with vector and graph databases such as pgvector, Neptune, or Memgraph. - Experience in a regulated field such as financial services or healthcare. - Familiarity with AWS SageMaker.