Backend Engineer, AI (Agent Systems)

About the role

We are looking for a Backend Engineer, AI to help build the infrastructure behind an AI-powered product and its agent-based workflows. In this role, you'll work on the layer connecting AI models with the product experience. You'll be responsible for building backend systems that make AI interactions fast, reliable, scalable and observable across mobile and desktop applications. The product relies on AI agents capable of handling multi-step tasks, maintaining context and interacting with external tools. This creates interesting engineering challenges around inference, orchestration, latency, reliability and cost.

What you'll do

• Build and operate backend services that power AI-driven product features in production
• Design inference pipelines and orchestration systems around AI models
• Define clear service boundaries and APIs connecting AI infrastructure with the rest of the product
• Take ownership of production reliability, including monitoring, logging, alerting and incident response
• Improve system performance by optimizing latency, throughput, caching, batching and streaming
• Work with AI/ML engineers to integrate models and inference capabilities into production systems
• Investigate and resolve issues in distributed systems, particularly under real production load
• Continuously improve backend architecture based on production data and real-world system behaviour

How we work

You'll join a small, high-talent-density and hands-on engineering team. The team moves quickly and makes decisions collaboratively, while maintaining a strong focus on engineering quality and continuous learning. Engineers are expected to bring structure to complex problems, exercise independent judgment and take ownership of delivering solutions. This is an environment for people who enjoy building ambitious products rather than simply maintaining the status quo. The ultimate goal is to make advanced AI genuinely useful in everyday life and deliver a product that can operate reliably at global scale.

Your work will directly contribute to the reliability and performance of the AI platform. Success in this role means: * Backend services reliably handle production AI workloads at scale * APIs remain stable, well-structured and easy to integrate with frontend and ML systems * AI infrastructure delivers strong latency and throughput * Production incidents are detected quickly, diagnosed efficiently and resolved with minimal impact on users * Monitoring and observability provide clear insight into system behaviour * Continuous improvements based on real-world usage lead to measurable gains in performance and reliability

What we're looking for

• Strong foundations in backend engineering and experience building production systems
• Experience developing or operating high-throughput, low-latency services
• Familiarity with AI inference technologies and patterns, including LLMs, embeddings or multimodal models
• The ability to troubleshoot and debug distributed systems under significant load
• A pragmatic, hands-on approach to engineering, with a strong focus on shipping and learning from real production behaviour
• The ability to work independently and make sound technical decisions in an environment where requirements can evolve quickly

Tech stack

• Python
• Node.js
• PyTorch
• OpenAI, Anthropic and open-source LLMs
• SQL / NoSQL
• Kubernetes
• Docker