Backend Engineer, AI

Backend Engineer, AI Location: Remote or onsite Employment type: Full-time About A1 A1 is building an AI assistant for everyday tasks across conversations, errands, organisation, and workflows. The product is designed to handle long-running workflows, maintain context across interactions, use external tools, and complete multi-step tasks with limited prompting. Our engineering work focuses on making these systems reliable enough for repeated, everyday use despite the non-deterministic behaviour of AI models. About the role We’re hiring a Backend Engineer, AI to build and operate the inference and orchestration layer behind A1’s AI features. You’ll work between models, backend services, and client applications, with responsibility for latency, reliability, observability, throughput, and infrastructure cost. Your systems will expose AI capabilities through production APIs used by mobile and desktop applications. What you’ll do Build and operate backend services for AI-powered product features. Design inference pipelines, orchestration systems, and service boundaries around AI models. Integrate LLMs, embedding models, multimodal models, and external tools into production workflows. Build APIs used by frontend and ML systems. Improve inference latency and throughput using caching, batching, streaming, and other optimisation techniques. Monitor production systems through logging, metrics, alerts, and tracing. Investigate incidents and failures across distributed services. Improve reliability based on production behaviour and user traffic. Make engineering trade-offs across performance, reliability, and infrastructure cost. What we’re looking for Production backend engineering experience. Experience building or operating high-throughput, low-latency services. Experience debugging distributed systems under production load. Working knowledge of APIs, databases, queues, caching, and service architecture. Familiarity with AI inference patterns including LLMs, embeddings, or multimodal models. Ability to diagnose performance and reliability problems using production data. Ability to take a backend system from design through deployment and operation. Nice to have Experience serving LLMs or other machine-learning models in production. Experience with model routing, tool calling, retrieval, or multi-step AI workflows. Experience optimising inference cost, token usage, caching, or model latency. Experience operating Kubernetes-based infrastructure. Experience building systems that stream model outputs to client applications. Tech stack Our current stack includes: Python Node.js PyTorch OpenAI, Anthropic, and open-source language models SQL and NoSQL databases Kubernetes Docker You do not need experience with every technology listed above. What a typical week may include You might spend your time: Building an API or orchestration service for a new AI feature. Investigating latency across model inference, tool calls, and backend services. Reviewing production traces to understand why a workflow failed. Improving caching, batching, or streaming to reduce response time or infrastructure cost. Working with frontend and ML engineers on service interfaces. Responding to production incidents and implementing fixes that prevent recurrence. Testing new models or inference approaches against production requirements. How we work A1 operates with a small engineering team where individual engineers own systems from design through production. Technical decisions are discussed as a team, while engineers are expected to make day-to-day implementation decisions independently. We release changes regularly, measure their behaviour in production, and use those results to decide what to improve next. Interview process If your experience appears relevant to the role, the process includes 3 interviews, with a maximum of 4 . Interviews may take place remotely or onsite and are conducted by members of the technical team. We assess candidates on backend engineering ability, system design, debugging, and their approach to building production AI systems. Hiring approach You do not need to match every item in this description to apply. We assess candidates based on the skills required to do the job and the experience they can bring to the team. We welcome applications regardless of gender, race, ethnicity, religion, disability, sexual orientation, age, or background. Show more Show less