Associate Principal - Research Engineering

We are looking for an Agentic AI Architect to design and deliver production grade agentic systems on AWS. This role owns the full lifecycle from architecture and design to hands-on implementation, bridging the gap between high level system design and working code. Design multiagent architectures, planning tool use, memory orchestration for real business use cases. Architect and deploy agentic workflows natively on AWS Bedrock, Lambda, Step Functions, SageMaker, ECS, EKS. Write production quality code in Python while architecting; this is a hands-on, build-as-you-design role.

Define agent-to-agent communication, tool/function-calling patterns, and orchestration frameworks (LangGraph, CrewAI, Bedrock Agents or custom). Build RAG pipelines and integrate vector stores (OpenSearch, Pinecone, etc.) into agent workflows. Establish guardrails, evaluation frameworks, and observability for agentic systems in production. Optimize for latency, cost, and reliability across distributed agent workflows on AWS. Collaborate closely with Python engineers, guiding them on implementation while co-building core agent components. Prototype rapidly, iterate based on evals, and take solutions from PoC to scaled deployment. Stay current with the agentic AI ecosystem and bring emerging patterns into the architecture.