Staff Software Engineer, Applied AI
Core
Build autonomous agent harnesses and evaluation systems for identity operations and security, enabling AI agents to safely investigate risk and complete tasks across enterprise systems.
Role type
Staff Software Engineer (Applied AI / Agent Systems)
Builds
Autonomous agent harnesses, evaluation datasets, and analytics layers for identity security operations.
Domain
Identity management, AI security, enterprise automation.
Deliverable
production ML models
Required skills
Python engineering, LLM application development, agent workflow design, model API integration, failure analysis, prompt engineering, security policy implementation, evaluation system design, asynchronous service integration, structured outputs.
Preferred skills
Experience with Claude Agent SDK, OpenAI Codex SDK, LangChain, LangGraph, Amazon Bedrock AgentCore, ClickHouse, BigQuery, fine-tuning, post-training evaluation, model distillation, open-source contributions.
Technologies
Python, FastAPI, AWS Step Functions, Amazon Bedrock, PostgreSQL, pgvector, OpenAI API, Anthropic API, ClickHouse, BigQuery.
Responsibilities
Build agent harnesses for planning, tool use, memory, and recovery; integrate OpenAI and Anthropic model APIs; evaluate existing harnesses and build missing pieces; maintain authorization checks and sensitive write policies; build evaluation datasets and replay tests; measure production metrics like false positives and completion rates; test for prompt injection and unsafe tool use; improve prompts and model routing.
Seniority
Staff, domain leadership and critical component implementation.
