Staff Engineer — Agentic AI

ABOUT THE ROLE

This is a senior technical leadership role at the heart of an AI-native engineering software company, owning the core agent intelligence layer that turns mechanical engineers' intent into reliable, cost-efficient multi-step workflows across complex desktop engineering tools. You'll report directly to the CTO and serve as the technical lead for a small team of AI engineers, a user researcher, and domain expert contractors. The work you do here will define real-world product value for enterprise customers.

WHAT YOU'LL DO

- Lead development of the agent intelligence layer that executes multi-step workflows across CAD, simulation, and PLM software.

- Own the full product loop — from user story definition to implementation to benchmarking against real engineering workflows.

- Drive agent task success rate by defining evaluation frameworks, establishing baselines, and iterating on performance.

- Set and enforce per-task token budgets and track cost per completed workflow to ensure commercial viability.

- Design rigorous, reproducible evaluation infrastructure grounded in validated user stories — think SWE-bench-level rigor applied to engineering workflows.

- Lead user story mapping and validation through direct interviews and collaboration with domain experts.

- Translate validated user stories into testable evals, closing the loop between user research and benchmarking.

- Own agent architecture decisions: tool-calling strategies, state management, error recovery, model routing, and context management.

- Act as a player-coach — write production code, review designs, unblock the team, and raise the engineering bar.

- Collaborate cross-functionally with integrations, product, and customers during POCs to align agent behavior with real-world usage.

WHAT WE'RE LOOKING FOR

- 7+ years in software engineering, including at least 2 years building and shipping real-world agentic LLM systems (tool calling, multi-step workflows, failure recovery, cost control).

- Deep experience with LLM application architecture: model selection, context/window management, retrieval strategies, tool-calling frameworks, and orchestration patterns.

- Strong evaluation and benchmarking instincts for agentic systems — task completion rates, cost efficiency, failure mode analysis; familiarity with benchmarks such as SWE-bench, GAIA, or τ-bench is a plus.

- Proven track record of shipped AI systems with measurable outcomes — not just demos or prototypes.

- Strong Python skills and hands-on familiarity with the LLM tooling ecosystem (function calling, tool use APIs, tracing/observability tools, evaluation frameworks).

- Technical leadership experience setting direction for small teams (3–6 engineers) and performing meaningful code review and architecture decisions.

- Hands-on background with mechanical engineering software — CAD/CAE/PLM or simulation tooling (e.g. Siemens NX/NXOpen, Teamcenter, CATIA, Creo, SolidWorks, Ansys, Abaqus, or similar) — either as a builder of these tools or as a power user inside an engineering or manufacturing org.

- Experience shipping AI/LLM tooling on top of proprietary engineering data or desktop engineering software (e.g. an agent or MCP server over CAD/PLM APIs, RAG over engineering repos or schematics).

- Familiarity with enterprise deployment constraints, including behavior on locked-down corporate workstations.

- Experience with desktop automation or programmatic control of applications (COM or similar) is a strong plus.

- Published work, open-source contributions, or benchmark contributions in agentic AI is a plus.

COMPENSATION & BENEFITS

Salary range: $160,000 – $250,000 USD annually. Visa sponsorship is not available for this role.

LOCATION

On-site in San Francisco, California, USA.