Senior Backend Engineer – LLM Inference

Role overview Build backend services and Kubernetes-native applications that make large language model inference reliable and accessible through a cloud platform. This is a hands-on senior role spanning design, implementation, deployment, and production operation, with close collaboration across infrastructure and AI engineering.

Responsibilities - Design, build, and operate production backend services in Go. - Develop Kubernetes-native applications and automation to support dependable platform operations. - Create APIs and integrations for inference services and related cloud capabilities. - Improve performance, scalability, and reliability through testing, observability, and operational changes. - Work with other engineering teams on technical design, delivery, and production troubleshooting. - Own work from initial design through rollout and contribute to technical decisions as the platform develops.

Requirements - Experience designing, shipping, and operating production backend services. - Strong production Go experience, including concurrency, context propagation, API design, and error handling. - Understanding of asynchronous systems, HTTP, and long-lived connections, including timeouts, cancellation, resource limits, and failure handling. - Distributed systems knowledge, such as caching, consistency, retries, idempotency, and reliable event processing. - Strong SQL and relational data modeling skills, including PostgreSQL, transactions, and schema migrations. - Practical experience building Kubernetes-native software, such as controllers or operators, custom resources, and reconciliation loops.

Nice to have - Rust, especially for asynchronous or network services; Python for integrating inference tooling. - Experience with API gateways, networking, traffic management, service discovery, GitOps, or deployments across clusters and regions. - Familiarity with LLM APIs or inference systems such as vLLM, SGLang, NVIDIA Dynamo, or llm-d. - Experience with Valkey/Redis, NATS JetStream, event processing, usage metering, or comparable state and messaging systems. - OpenTelemetry, Prometheus, or capacity planning for workloads with variable request costs.

Benefits and work setup - Senior, full-time permanent role based in Helsinki or London, or remote in Europe. - Cash and equity compensation, plus fringe benefits.