Senior Software Developer: Models Team (Token Factory)

About Nebius: 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. Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D. About the Product   Token Factory is focused on building a next-generation platform that enables companies to seamlessly integrate AI into their products and workflows. Our vision is to create a powerful, open, and scalable alternative for deploying and managing AI systems—making advanced AI infrastructure more accessible to both fast-growing startups and large enterprises.   We work with a wide range of customers, from AI-first companies to established technology organisations, helping them run AI workloads reliably at scale. Our goal is to become a leading platform for high-performance AI inference, delivering predictable latency, strong reliability, and the ability to scale to meet demanding production needs.   Customer feedback plays a central role in how we build—our development process is highly iterative and closely aligned with real-world use cases.   About the Team The Models Team is responsible for onboarding state-of-the-art (SOTA) open-source models into Nebius TokenFactory, including models such as DeepSeek V4 Pro, GLM 5.1, Kimi K2.6, and Minimax M2.7. A major focus of the team is serving large-scale AI models efficiently and reliably in production. To achieve this, we work on advanced inference and systems optimization techniques, including: Cache-aware routing NUMA-aware deployments KV-cache offloading Disaggregated serving architectures Autoscaling with high-speed model loading over InfiniBand / RoCE The team maintains and extends forks of leading inference frameworks such as vLLM and TRT-LLM . We have deep expertise in production-scale model serving and regularly support the Solutions Architects team on the most demanding customer PoCs. To operate efficiently at scale, we invest heavily in tooling and automation.  Examples include: Performance, quality, and smoke-testing frameworks Hyperparameter optimization for inference framework configurations Gibberish detection systems Automated rollout pipelines for inference framework upgrades Diagnostics and observability tooling Traffic replay systems Automated search for optimal serverless deployment configurations We collaborate closely with model builders, open-source communities, Nebius Cloud teams, and hardware vendors to continuously improve our serving infrastructure. The team is highly goal-oriented and outcome-driven, with a strong focus on delivering results rather than following rigid processes. Team Structure We are currently a team of eight engineers distributed across Europe, with members based in the Netherlands, the United Kingdom, Germany, and Latvia. Our workflows are optimized for remote collaboration. At the same time, we meet in person every one to two months at one of our locations to work together, brainstorm new ideas, and plan upcoming milestones. While many team members joined without extensive AI/ML experience, we have rapidly developed strong expertise in large-scale model serving and AI infrastructure. Technology Our work is deeply integrated with the broader cloud and infrastructure ecosystem. We primarily use Go and Python to build and scale backend systems. We collaborate closely with teams working on cloud infrastructure, observability, reliability, fault tolerance, and platform engineering. The challenges we solve sit at the intersection of distributed systems, high-performance computing, and modern AI infrastructure. We expect you to have: Experience serving LLMs in production Strong Python and/or Go programming skills Experience designing and operating highly scalable, highly available distributed services Nice to have: Contributions to vLLM, SGLang, TRT-LLM, or NVIDIA ecosystem open-source projects Deep understanding of KV cache management, speculative decoding, and quantization Experience with LLM evaluation frameworks Hands-on experience with performance benchmarking and optimization Deep understanding of Kubernetes Familiarity with distributed serving architectures and autoscaling Knowledge of InfiniBand, RoCE, or high-performance networking Benefits & Perks: Competitive compensation Career growth and learning opportunities Flexibility and ownership Collaborative and innovativ