Senior/Staff Software Engineer, Agentic Platform

Recruiting From Scratch•San Francisco, CA

1d•Hybrid

About The Position

The Senior/Staff Software Engineer, Agentic Platform will help shape a new agentic engineering team and influence the architecture and technical direction of a core AI product area. The company is building a platform that helps large manufacturers improve product reliability and reduce the cost of poor quality by connecting fragmented operational data from customer, service, field, telematics, warranty, and aftermarket systems. Its agentic platform turns complex, unstructured real-world data into actionable insights for engineering and manufacturing teams. The role involves designing and scaling agentic systems and the distributed infrastructure that carries customer data from ingestion through agent-driven reasoning to root-cause insights. The engineer will build multi-step and multi-agent workflows, handle noisy and heterogeneous data, and establish rigorous evaluation, observability, and reliability practices. Because agent conclusions may influence product redesigns, quality investigations, and recalls, accuracy and reliability are central to the work. This is a hands-on technical leadership role. The engineer will partner with Engineering and Product leadership on architecture and roadmap decisions, lead technical discussions, set engineering standards, and mentor other engineers while remaining closely involved in implementation. The position is hybrid in San Francisco, California, and offers a base salary of $220,000–$290,000 plus equity.

Requirements

At least 4 years of professional software engineering experience.
Recent, hands-on experience building production agentic systems.
Experience with LLM orchestration, tool or function calling, multi-step reasoning, and multi-agent systems.
Understanding of retrieval, context engineering, memory, and MCP or comparable agent integration protocols.
Experience designing, building, and owning complex distributed systems.
Strong understanding of the modern agentic technology stack and infrastructure needed to operate LLM-driven systems reliably.
Experience building evaluation, testing, and observability frameworks for non-deterministic AI systems.
Ability to work with messy, heterogeneous, and unstructured real-world data and make it usable for AI-driven reasoning.
Strong system design and architectural decision-making skills.
Experience providing technical mentorship and guidance without requiring formal management authority.
Ability to collaborate with Engineering and Product leadership to shape technical direction.
Strong ownership mentality and ability to lead complex technical initiatives from design through production.
Ability to independently own complex architecture and infrastructure decisions while remaining hands-on.
Ability to work effectively in a hybrid role located in San Francisco, California.

Nice To Haves

Experience in data-intensive B2B SaaS environments.
Experience building production agentic platforms or AI-powered enterprise products.
Experience with multi-agent orchestration and complex reasoning workflows.
Experience building AI systems that operate on large, fragmented enterprise datasets.
Experience working in manufacturing, industrial technology, automotive, or other complex enterprise environments.
Experience building root-cause analysis, anomaly detection, or operational intelligence systems.
Experience with LLM evaluation frameworks, observability tooling, and production monitoring.
Experience designing systems where incorrect AI outputs can have significant operational or financial consequences.
Experience leading architecture discussions and influencing engineering direction.
Experience mentoring engineers and raising technical standards.

Responsibilities

Design and scale core agentic systems, including orchestration, tool use, memory, retrieval, context engineering, evaluation, observability, and reliability guardrails.
Architect and develop distributed infrastructure that processes complex customer data from ingestion through agent-driven reasoning to actionable root-cause insights.
Transform fragmented, noisy, incomplete, and heterogeneous operational data into structured context that AI models can use reliably.
Build and harden multi-step and multi-agent workflows that reason across large volumes of enterprise data.
Develop rigorous testing and evaluation processes for non-deterministic systems, measuring agent accuracy against expert-validated ground truth.
Evaluate system performance through customer outcomes such as detection rates, investigation speed, and warranty cost reduction.
Partner with Engineering and Product leadership to define the agentic platform's architecture, technical vision, and roadmap.
Lead technical discussions, contribute to design documents, conduct code reviews, and establish best practices for reliable agentic systems.
Provide technical guidance and mentorship, particularly in agentic architecture, distributed systems, and system design.