Staff Engineer, Agentic AI

About the Role

Lead the development of agent intelligence that helps mechanical engineers complete multi-step workflows in desktop engineering software. You will combine hands-on engineering, technical leadership, and collaboration with user research and domain experts to improve real-world task completion and product value.

What You'll Do

• Own agent task success metrics, establish baselines, and systematically improve completion rates across engineering workflows.

• Design and build reproducible evaluation infrastructure based on validated user stories, including tests for cost efficiency and failure modes.

• Set per-task token budgets and track the cost of completed workflows.

• Interview engineers with research and domain teammates, document their workflows, and translate validated stories into benchmarks.

• Prioritize workflow coverage based on customer value and technical feasibility.

• Set the agent architecture for tool calling, state management, error recovery, model routing, and context management.

• Lead a small AI engineering team while writing production code, reviewing designs, and unblocking teammates.

• Work with product, integrations, and customers during proof-of-concept projects to align agent behavior with real usage.

What We're Looking For

• At least 7 years of software engineering experience, including 2 or more years building LLM-based agents that take real-world actions.

• Strong Python skills and deep experience with LLM application architecture, including model selection, retrieval, context management, orchestration, and tool calling.

• Experience building rigorous agent evaluation and benchmarking systems that measure task completion, cost, and failure modes.

• Technical leadership experience on a small engineering team, with a track record of reviewing code and continuing to contribute production software.

• Experience shipping AI or LLM tools using proprietary engineering data or desktop engineering software.

• Familiarity with function calling, tool APIs, tracing or observability, and evaluation tools.

• Experience with desktop automation or application APIs, such as COM, and a background in mechanical engineering software, CAD/CAE, PLM, or an adjacent domain.

• Understanding of deployment constraints on managed enterprise workstations.

Compensation & Benefits

Base salary: $160,000 to $250,000 USD annually. Visa sponsorship is not available.

Location

On-site in San Francisco, California, United States.