AI - Principal Software Engineer

Cohu•San Diego, CA

1d•$175,000 - $225,000•Hybrid

About The Position

This is an opportunity for an AI-Engineering Principal Software Engineer, AI Group to get in on the ground floor. Cohu is building the AI engineering platform for semiconductor test from the ground up: cloud and on-prem inference infrastructure, custom agent tooling, and the software integrations that let AI-driven workflows reach application engineers. As both a hands-on engineer and technical leader, this role ships that platform hands-on, sets its technical direction, and leads the small team building it. This position will own technical direction for one of AI-Engineering’s teams: architecture calls, day-to-day prioritization, and the long-term platform viewpoint. You will be defining practice as much as following it, and your team’s goals will shift as the group learns what customers and field engineers actually need.

Requirements

Bachelors Degree in Computer Science or equivalent; EE/physics acceptable with a strong software track record.
5+ years relevant experience, prior technical-lead or architect experience preferred.
Senior-level ability to design, build, and ship production software independently.
Reach for AI coding agents (Claude Code, Cursor, Codex, Antigravity, or similar) by default, not because a policy directs you to. Focus on what the agents can do, have opinions about using them well, and make the call on what the team standardizes.
Demonstrated, senior-level ability to drive cross-team technical decisions to closure and gain buy-in without formal authority.
In-depth experience architecting distributed systems in Kubernetes/cloud environments, including production ownership.
Frontend/UI skills sufficient to ship a real agent-facing application, not just a demo.
Strong individual and small team leadership skills in prioritizing and shipping.
Excellent written and verbal communication.
Comfortable presenting technical direction to cross-functional and leadership stakeholders.
Working fluency with LLM/agent tooling: agent frameworks, MCP or equivalent toolintegration patterns, RAG components preferred.
C++ and Python knowledge and prior production experience in either is a plus.
Hands-on experience selecting, installing, and optimizing GPU/server hardware and inference stacks (e.g., vLLM, TensorRT-LLM), ideally including air-gapped/on-prem environments.
Semiconductor industry familiarity, particularly test/ATE domain.

Nice To Haves

prior technical-lead or architect experience preferred
C++ and Python knowledge and prior production experience in either is a plus.
Semiconductor industry familiarity, particularly test/ATE domain.

Responsibilities

Personally design, build, and ship agent tooling, RAG systems, or infrastructure work alongside directing others.
Set technical direction for software integration, cloud infrastructure, and on-prem hardware selection/deployment for air-gapped LLM inference: server/GPU selection, inference software stack, model evaluation methodology.
Architect and drive team-wide adoption of reusable patterns for custom AI agent development (workflows, tool integration, hardware integration), RAG/retrieval architecture, and custom UIs built on top of them.
Responsible for the Kubernetes-based infrastructure, your systems run on: capacity planning, resource budgeting, scaling, namespace/cluster lifecycle, production incident response.
Provide guidance, coaching, and day-to-day prioritization for the team. Participate in architectural planning and represent the team’s long-term viewpoint.
Serve as technical escalation point for architecture calls that cross team or service boundaries.
Potential to act as AI Engineering lead.
Stay ahead of the model-serving and inference-hardware landscape.