Senior/Staff Software Engineer, Agentic Platform

About the Company

An AI platform helping manufacturers cut the cost of poor quality and improve product reliability, building out a growing agentic platform team in San Francisco alongside an existing engineering group in New York.

About the Role

This is a senior-to-staff individual contributor seat owning the agentic platform that turns messy, disconnected manufacturing data, including customer, service, field, telematics, warranty, and aftermarket signals, into reliable, production-grade reasoning. The team is fast-moving and high-accountability, with direct feedback and low ego, and the company is candid that there are no real experts in this space yet: experience with experiments that failed is valued as highly as experience with ones that succeeded.

What You'll Own

• Technical architecture for agentic systems, including orchestration, tool use, memory, retrieval, evaluation, observability, and guardrails for noisy and incomplete input data
• The reasoning layer and infrastructure that take customer data from ingestion through agent-driven reasoning to root-cause insight, at manufacturer scale and reliability
• Rigorous evaluation and testing for non-deterministic systems, measuring agent accuracy against expert-validated ground truth and outcomes like detection rates, investigation speed, and warranty cost
• Partnership across Engineering and Product to shape the technical vision and roadmap for the agent platform
• Raising engineering standards through code review, design docs, and technical direction
• Mentoring engineers on agentic patterns and system design, unblocking hard problems and sharing context, without formal management authority

What We're Looking For

• 4+ years of engineering experience, senior to staff level, not early-career
• Recent, hands-on production work on agentic systems: orchestration, tool use, multi-agent reasoning, retrieval, and evaluation, ideally with at least the last year including agentic experience and a production launch
• Enterprise-scale experience, not a fully consumer-only background, ideally from a large-scale AI platform or cloud AI/ML infrastructure team (comparable to teams at Amazon Bedrock, Google Vertex, or Meta AI)
• Real intuition for non-deterministic failure: can design and own complex, data-intensive distributed systems, ideally in B2B or enterprise SaaS environments
• A data-first philosophy with sound judgment in ambiguous, fast-moving situations
• Stable tenure, not a pattern of six-to-nine-month moves
• A strong executor who can balance shipping MVPs with a real commitment to production-grade systems
• Bonus: production scars from agentic work you can speak to in detail, with strong opinions on what you built and why

Logistics

San Francisco, hybrid. Compensation is $220K to $290K base plus equity.  Visa sponsorship is not available for this role, and US work authorization is required.