AI Platform Engineer

The role

You will build the execution and evaluation layers for Supabase's internal AI systems. This includes building an agent platform, enforcing governance in code, and designing how the system contacts people.

What you will do

• Ship the agent platform to production
• Own the evaluation layer and the gate that depends on it
• Build and register the agent portfolio
• Enforce governance in code
• Design how the system contacts people
• Own the platform tooling
• Compute the operating measures from production data
• Instrument the platform's own return

Required qualifications

• Have shipped production LLM agent systems that other people depended on
• Design evaluations, not spot checks
• Have done deep API work against the systems work actually lives in
• Authored MCP servers
• Own infrastructure end to end in Python on GCP, with a cloud warehouse and infrastructure as code
• Have taste about how software contacts humans

Nice to have

• Public work in this space
• LLM observability and cost instrumentation
• Built an internal platform that non-engineers adopted voluntarily

Benefits

• WeWork membership or co-working allowance
• ESOP (equity ownership)
• Tech Allowance
• 100% health insurance for employees
• 80% health insurance for dependents
• Annual Off-Sites
• Flexible Work
• Annual education allowance

About the company

Supabase is the Postgres development platform, built by developers for developers. They provide a complete backend solution including Database, Auth, Storage, Edge Functions, Realtime, and Vector Search.