Backend Engineer, AI (Agent Systems)

Role overview Own the backend inference and orchestration systems that connect AI models to product experiences. You’ll build and operate production services for AI features across mobile and desktop, where response time, correctness, reliability, and cost all affect how well the product works.

Responsibilities - Build and operate backend services that deliver AI-powered features in production. - Design inference pipelines, orchestration layers, and service boundaries around models. - Monitor service health and handle logging, alerting, and incident response. - Improve latency and throughput through techniques such as caching, batching, and streaming. - Keep APIs stable and clear for integration with client, ML, and product systems. - Diagnose production problems and use observed usage to guide reliability and performance improvements.

Requirements - Strong backend engineering fundamentals in production environments. - Experience operating high-throughput, low-latency services. - Familiarity with AI inference patterns, including language models, embeddings, or multimodal systems. - Ability to debug distributed systems under load. - A practical, iterative approach to shipping and learning from production behavior.

Nice to have The stated environment includes Python, Node.js, PyTorch, commercial and open-source language models, SQL and NoSQL databases, Kubernetes, and Docker.

Benefits and work setup The team describes a hands-on environment with collective decisions, fast iteration, and independent ownership. The interview process typically involves three or four conversations, held virtually and/or onsite.