Senior AI Engineer

Job Overview

We're hiring a Senior AI Engineer to design, build, and operate intelligent agent systems end-to-end on our AWS-native GenAI platform. You'll own major agent services from design through stable operations, design agent workflows and RAG architectures for team-level use, define testing and reliability practices for your areas, and mentor across the team. Agents are the headline, but this is a full-stack platform role: you ship the FastAPI service, the Terraform, the MCP server, and the eval harness behind the agent — and you set the bar for how they're built.

Essential Duties & Responsibilities

• Design and own LLM-powered agents end-to-end for internal tooling, customer-facing features, and workflow automation — built on Bedrock with LangChain/LangGraph and the langchain-aws/-anthropic/-core stack
• Design agent workflows and RAG architectures for team-level use; influence team design beyond your own services
• Build and operate MCP servers that expose company data and services as structured tools for AI models
• Define the testing, reliability, and operational practices for your areas; own delivery and on-call quality for major services
• Stand up evaluation pipelines (Langfuse + eval harnesses) and prompt-versioning practices, and use them to drive reliability and cost/quality improvements
• Lead design reviews, mentor across the team, and drive improvements in engineering practice
• Collaborate with product, data, and platform teams to scope and deliver agent-based solutions, applying platform patterns and governance hooks correctly

Requirements & Skills

• 5+ years of software engineering experience, with multiple years building and operating LLM or AI agent systems in production
• Expert-level Python and service development (async, FastAPI, Pydantic v2), with a track record of owning services end-to-end
• Deep GenAI/LLM application experience — designing RAG architectures and agent workflows, ideally with Claude via Amazon Bedrock or the Anthropic API
• Strong AWS engineering (Bedrock, Lambda, ECS, API Gateway, and the surrounding services)
• Hands-on agent tooling: LangChain/LangGraph and MCP (building and operating tool/resource servers)
• Terraform / IaC and containerized delivery through CI/CD
• Strong data-layer experience (Redis, Snowflake, DynamoDB, Postgres/Aurora) via async SQLAlchemy
• Identity and app-security fundamentals (JWT/OAuth, secrets management)
• Production observability and LLM evaluation experience (tracing, evals, latency/cost/quality metrics)
• Demonstrated ability to drive improvements in engineering practice and mentor team members