Backend Engineer, AI (Agent Systems)

Role overview Own the backend inference and orchestration layer that connects AI models to product users. The work is focused on production systems for long-running workflows and persistent context, where API stability, response time, correctness, reliability, and cost shape the user experience across mobile and desktop clients.

Responsibilities - Build and operate backend services that support AI-powered product features. - Design inference pipelines, orchestration logic, and service boundaries around models. - Make production behavior observable through monitoring, logging, and alerting, and take part in incident response. - Improve latency and throughput through approaches such as caching, batching, and streaming. - Diagnose distributed-system issues under load and resolve their underlying causes. - Use real production behavior to guide iterative improvements to performance and reliability. - Provide clear, stable APIs that integrate with client and machine learning systems.

Requirements - Strong backend engineering fundamentals from production environments. - Experience operating services that need high throughput and low latency. - Familiarity with AI inference patterns, including language models, embeddings, or multimodal systems. - Comfort debugging distributed systems under load. - Ability to ship, learn from production behavior, and work independently through ambiguous problems.

Nice to have The listed environment includes Python, Node.js, PyTorch, SQL and NoSQL databases, containers and orchestration, and hosted or open-source language models. The team describes a hands-on, fast-moving approach with collective decisions and a balance between delivery and learning.