Software engineer in serverless AI

Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure. We're looking for a Senior Software Engineer to help us build Nebius Serverless AI — our GPU-native platform for deploying inference endpoints, batch jobs, and AI workloads without managing infrastructure.

• Design and build core components of the Serverless platform — the control plane, scheduler, runtime, autoscaler, and the APIs customers integrate against.
• Own the hardest engineering problems on the service: cold-start latency, GPU scheduling under contention, multi-tenant isolation, fair-share quotas, request routing at the edge.
• Set the technical direction for the areas you own — drive architecture decisions, write the design docs, and align the team behind the chosen approach.
• Raise the engineering bar across the team through code reviews, design reviews, and example — the kind of senior presence that compounds.
• Run the service like an SRE: define SLOs, build observability, lead incident response, and make sure every postmortem changes the platform, not just the runbook.
• Work directly with customers on architecture reviews, performance escalations, and the production problems that don't fit a ticket template.
• Partner closely with Product, GTM, and infrastructure teams to translate customer needs into a coherent technical roadmap.

• 7+ years of professional software engineering experience, with a track record of shipping production distributed systems at scale.
• Excellent knowledge of Golang, or you are ready to quickly switch to it.
• Deep experience with Kubernetes and container orchestration — you've operated it in anger, not just deployed it.
• Strong distributed-systems instincts: consistency vs. availability trade-offs, queueing, backpressure, retries, idempotency, multi-tenancy.
• Experience designing and operating high-throughput, low-latency services — you know where the milliseconds go and how to get them back.
• A history of being the engineer others look to on hard problems — driving design discussions, unblocking teammates, and shipping the thing nobody else wanted to touch.
• Ability to write reliable code and dig into complex problems.
• Teamwork-oriented approach.
• Будет плюсом: Experience building serverless or function-as-a-service platforms (Knative, AWS Lambda, GCP Cloud Run, Cloudflare Workers, Modal, Replicate, Together, Fireworks, Anyscale, or similar); GPU scheduling experience — Kubernetes device plugins, MIG, MPS, time-slicing, NVIDIA GPU Operator; ML inference experience — vLLM, TensorRT-LLM, Triton Inference Server, SGLang, model loading and warm-pool strategies; Cold-start optimization at the runtime, image, or snapshot level (FireCracker, gVisor, checkpoint/restore, image streaming); Experience writing Kubernetes operators (Go + controller-runtime / kubebuilder); Contributions to relevant open-source projects in the serverless, scheduling, or inference ecosystems

• You'll work from our office in Amsterdam or London, hybrid.
• Competitive compensation
• Career growth and learning opportunities
• Flexibility and ownership
• Collaborative and innovative culture
• Opportunity to work on impactful AI projects
• International environment and talented teams
• Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI