Founding Backend Engineer

About the Role

As a founding backend engineer at an early-stage AI assistant company, you will own the reliable systems that power an AI agent communicating through text, email, and phone. This hands-on role spans designing, building, operating, and improving the backend infrastructure that keeps messages and scheduled actions working dependably.

What You'll Do

• Design and operate messaging pipelines for inbound and outbound communications, with reliable ordering, retries, and delivery.
• Build the background agent runtime, including concurrency, timeouts, cancellation, and recovery from model or tool failures.
• Develop scheduling systems that execute deferred actions reliably and in the correct time zone.
• Use capacity planning, load testing, database optimization, and architectural improvements to prepare systems for growth.
• Maintain secure data practices, including credential encryption, Postgres schema design and migrations, and privacy-related deletion workflows.
• Build observability and cost tracking that trace messages through downstream model and tool calls.
• Set backend engineering standards and help recruit and mentor future team members.

What We're Looking For

• At least 4 years of software engineering experience, including substantial production on-call responsibility.
• Deep distributed-systems experience with queues, workers, retries, idempotency, and production failure modes.
• Experience scaling production systems through rapid growth and making sound architectural tradeoffs.
• Production experience with Postgres, including schema design, migrations, and performance tuning, as well as cloud infrastructure and infrastructure as code.
• Production experience with TypeScript or Python.
• Experience building messaging, telephony, or event-driven systems where dropped or duplicated events have meaningful consequences.
• Experience with observability, tracing, capacity planning, and measuring latency and cost across distributed systems.
• A hands-on approach to technical leadership, plus the ability to verify and own code produced with AI-assisted development tools.