Applied AI Engineer, Agent Systems

The role

You will build the application layer between a foundation model and users, including client context layers, agent systems, and insight pipelines. You will start embedded in the field with a Deployment Lead to ship tools for an orchard, then convert these into reusable protocols.

What you will do

• Build client context layer and retrieval
• Develop tool schema exposing platform data and insights
• Build product agent systems including skill and instruction files
• Build durable, idempotent, and observable scheduled or event-driven work
• Create one insight pipeline rendering to chat, email, PDF, and API
• Develop context schema and card structure for multi-tenant isolation and IP protection
• Ensure reliable actuation of client-set policies
• Manage eval suites, trace capture, drift monitoring, and per-client cost accounting

Required qualifications

• Has built and measured production AI retrieval and context assembly
• Pushes repeatable work into deterministic code
• Has shipped multi-tenant isolation in production
• Has built a config or skill system used by non-engineers
• Has operated an agent system in production
• Has built durable, idempotent, observable scheduled or event-driven agent work
• Has built eval suites, trace capture, or drift monitoring
• Has a full-stack or front-end web/application engineering background

Nice to have

• Integration grit against client and legacy systems: enterprise APIs, CSV over SFTP, email deliverability

Benefits

• Standard benefits

About the company

The Electric Plant Co. is building a new category of plant and tree intelligence using IoT hardware and a foundation model called Bombadil. They are a small, fast-moving company in San Francisco working at the intersection of biology, hardware, AI, and IoT.