Software Engineer - AI Tooling Infrastructure

Responsibilities

• Build and operate production backend services for Supabase’s AI products, including deployment, scaling, configuration, and releases for the MCP server.
• Own service architecture and evolve or split services to improve safety, reliability, and maintainability.
• Define and meet reliability targets with SLOs, error budgets, metrics, alerts, and runbooks.
• Participate in the on-call rotation, respond to incidents, coordinate across teams, and document follow-ups.
• Support the team queue and convert recurring issues into durable fixes.
• Own authentication and authorization for AI and agent clients, including OAuth applications, scoped tokens, permissions, and backward-compatible permission changes.
• Own observability through metrics, traces, logs, dashboards, and alerts.
• Manage agent-traffic capacity and abuse controls, including rate limits, cost controls, safety controls, and dependency degradation behavior.
• Collaborate with Platform and SRE teams on shared code, releases, infrastructure, alerting, and incident practices.
• Contribute to AI Tooling code where it affects reliability and security.

Requirements

• 5+ years of experience building and operating backend services in production, including on-call responsibility.
• Strong TypeScript or Go skills and comfort working with the other language.
• Experience designing and operating APIs with real consumers, including versioning, error handling, rate limiting, and backward-compatible changes.
• Strong production knowledge of OAuth, OIDC, JWT, and API authorization.
• Good understanding of PostgreSQL and relational databases.
• Comfort with CI/CD, Docker, containerized deployments, and infrastructure as code.
• Ability to debug requests across services using metrics, traces, and logs.
• Strong written communication and ability to work with distributed stakeholders in an asynchronous environment.
• Deep expertise in at least one of APIs and authorization, production operations, or gateways and distributed systems.
• Experience with agent tooling such as MCP, AI-agent or LLM-client traffic, service extraction, Grafana, OpenTelemetry, Logflare, SLOs, or open-source contributions is preferred.

Benefits

• Fully remote work with global hiring and no Supabase offices.
• WeWork membership or coworking allowance.
• ESOP equity ownership for every team member.
• Technology allowance for work equipment.
• Health insurance coverage of 100% for employees and 80% for dependents.
• Annual company off-site in a new city.
• Flexible asynchronous work arrangements.
• Annual professional development and education allowance.