Principal Software Engineer, Applied AI

What you'll do

• Build agent harnesses for planning, tool use, memory, approvals, interruption, and recovery; integrate them with workflow execution.
• Develop evaluation datasets, replay tests, and measurements; investigate production failures and turn them into regression tests.
• Protect authorization and tenant boundaries, verify completed actions, and improve prompts, retrieval, tools, and model routing.

What they ask for

• Strong Python engineering skills and experience shipping an LLM application that uses tools.
• Hands-on experience building agent workflows or evaluation systems for tool-using LLM applications.
• Experience integrating model APIs, structured outputs, and asynchronous services.
• Depth in agent execution or evaluation systems, including working code and failure analysis.

Nice to have

• Open-source contributions, agent observability, security evaluations, or synthetic evaluation data.
• Fine-tuning or post-training experience; identity and security workflows.

Pay

$224,000—$308,000 USD annual base salary

Tools

Python FastAPI, AWS Step Functions, Amazon Bedrock AgentCore, PostgreSQL on Amazon RDS, pgvector, OpenAI, Anthropic, ClickHouse or BigQuery

AI focus

Build autonomous agents for identity operations and security; develop agent harnesses and evaluation systems, integrate model APIs, and assess fine-tuning or other post-training methods when justified and approved.