The RoleYou'll own Bedrock's agentic AI platform: the infrastructure, frameworks, and tooling that power AI agents across the company. You'll also build and ship the first wave of production agents: from agents that help engineers understand our autonomous systems, to Field Triage agents that accelerate debugging of real-world issues, to workflow automation agents that keep our distributed team aligned.
This is a high-impact, high-autonomy role at the intersection of AI/LLM engineering, cloud infrastructure, and developer tooling. Part of the Cloud Platform team, you will work closely with the Autonomy team, and the Product/Ops teams to make agents a core part of how Bedrock operates.
What you'll do- Build the agentic platform that enables Bedrock engineers to create, deploy, and monitor AI agents on AWS. This includes agent orchestration, tool/MCP server integration, evaluation and observability, and production deployment pipelines.
- Ship the first agents grounded in Bedrock specificities. These include a Q&A agent for understanding system behaviors, a Field Triage agent for mapping real-world issues to spec nodes, and an Annotator agent that improves data labeling quality.
- Build workflow automation agents for internal productivity-Slack thread summarization, Linear ticket management, cross-tool orchestration-that make every team at Bedrock faster.
- Design and implement MCP servers that expose Bedrock's internal systems as tool interfaces for agents.
- Build evaluation and observability infrastructure for agents-tracing, cost tracking, quality metrics, and automated regression testing to ensure agents are reliable and improving.
- Iterate on agent UX patterns beyond simple chat-building interfaces that surface agent outputs in the right context (Slack, dashboards, CLI, pull requests) and enable human-in-the-loop workflows.
Also, you get to drive 100,000 lb excavators.
What we're looking for- 4+ years of professional experience in software engineering, with demonstrated ownership of production systems.
- Strong Python skills and comfort with async patterns, API design, and cloud-native development on AWS.
- Hands-on experience building LLM-powered applications including agents, tool use, RAG, prompt engineering, and evaluation. You've shipped something real with LLMs, not just experimented.
- Systems thinking. You can design an agent platform that's modular, observable, and maintainable, not just a collection of scripts.
- Comfort working across the stack. You'll interface with cloud infrastructure, data systems, frontend teams, and end users and you're energized by that breadth.
- Strong written communication. Agents are only as good as their grounding context. You'll help shape how we encode domain knowledge for AI consumption.
Preferred Qualifications- Experience with agent frameworks such as AWS Strands, LangGraph, CrewAI, Claude Agent SDK, or similar orchestration tools.
- Familiarity with MCP (Model Context Protocol) or similar tool-use / function-calling patterns for LLMs.
- Experience building developer platforms or internal tooling that other engineers use daily.
- Experience with observability and evaluation frameworks for AI/ML systems (tracing, cost analysis, quality metrics).
- Experience in an early-stage startup environment designing, building, and launching new products or major features from scratch.
Our roles are often flexible. If you don't fit all the criteria, or are in another location (especially one where we have an office like SF or NY) please apply anyway! We'd love to consider you.
