Software Engineer, AI

What you'll do

• Build and maintain end-to-end agent capabilities including tools, workflows, prompts, and partner system integrations.
• Develop the agent runtime for multi-agent orchestration, context memory, prompt caching, and model routing.
• Optimize system performance for p50 and p99 metrics, bursty traffic, and conversation cost efficiency.
• Create eval harnesses and LLM judges to measure model and prompt changes before deployment.
• Build assistants for merchant staff using private data and browser agents for non-API environments.
• Develop a unified conversation platform for real-time streaming across SMS, iMessage, email, and in-app chat.
• Implement authentication and fine-grained authorization to secure agent access to member and merchant data.
• Own the full lifecycle of backend services from initial design and coding to production rollout and on-call support.

What we're looking for

• 4+ years of experience building and operating backend services in production.
• Proficiency in at least one backend language and experience running services in the cloud.
• Solid distributed systems fundamentals including async processing, pub/sub, idempotency, retries, and caching.
• Hands-on experience with LLMs and a curiosity for making them reliable.
• Ability to demonstrate product judgment by identifying issues in prompts, tools, data, or services.
• Experience with daily use of AI coding agents and the judgment to evaluate their output.
• Ability to produce clear writing for design docs, PR descriptions, and agent prompts.
• Experience shipping LLM-powered features, using agent frameworks, or managing model inference (preferred).