Software Engineer, Applied AI

Role overview

Build software agents that automate real operational work in electronics manufacturing, from inspecting circuit board images to supporting factory operations and back-office processes. You’ll find manual workflows, decide which parts need model judgment and which are better handled with conventional code, then build and evaluate an end-to-end solution. The role also includes improving the shared platform used to run agents and working closely with the people who use their outputs.

Responsibilities

- Automate workflows such as visual inspection, machine setup, factory monitoring, invoice processing, order follow-up, and issue triage. - Design agent tools and supporting software, separating model decisions from deterministic checks and actions. - Build evaluations using human judgments, historical cases, and shadow runs before allowing systems to operate independently. - Track accuracy and operating cost, and retest changes to models, prompts, or configurations. - Improve agent infrastructure and reliability, including safe execution and protection against duplicate business actions. - Collaborate with factory and business teams to make systems more accurate, efficient, and useful.

Requirements

- Experience putting LLM agents into production, including understanding failures and measuring whether systems work. - Strong software engineering skills in areas such as concurrency, idempotency, data modeling, and security. - A practice of building evaluations before trusting automated results. - Interest in applying software to physical systems such as cameras, robotics, and manufacturing equipment. - Familiarity with the listed stack: TypeScript, Python, PostgreSQL, cloud infrastructure, and agent APIs and tools.

Nice to have

Experience with computer vision, optimization, or electronics.

Benefits and work setup

Compensation is listed as £80,000–£125,000 in the UK or $125,000–$250,000 in the U.S., depending on experience, plus equity.