Top Tech Jobs & Startup Jobs

Develop and optimize GPU clusters and InfiniBand networks for hyperscale HPC environments. Responsibilities include troubleshooting hardware and infrastructure issues, integrating new GPUs through Kubernetes, QEMU, and KVM, managing GPU devices and InfiniBand fabrics, tuning performance, and improving automated monitoring and fault resolution.

Conduct frontier applied research on efficient AI model architectures, including sparse attention, long-context memory, selective computation, dynamic inference, reasoning, and continual adaptation. Formulate research questions, design rigorous experiments at meaningful scales, develop efficient implementations, publish findings, contribute to open-source tools and models, mentor researchers, and help direct the research stream.

Develop and manage AI and ISV partner business development initiatives for Nebius’s AI cloud platform. The role involves collaborating with partners, managing strategic relationships, and supporting growth across AI infrastructure and cloud services. The posting provides limited specific responsibilities or qualifications, but emphasizes English proficiency and authorization to work in the country of application.

Own optimization of LLM and VLM inference systems from model artifacts through production deployment. Improve latency, throughput, memory efficiency, GPU utilization, quality, reliability, and cost per token. Deploy and extend inference engines, build model-compression workflows, implement decoding and KV-cache optimizations, create benchmark harnesses, and diagnose bottlenecks across models, kernels, runtimes, schedulers, gateways, and clusters. Collaborate with research, infrastructure, product, kernel, and customer teams.

Own production optimization of LLM and VLM inference systems, improving latency, throughput, memory efficiency, GPU utilization, quality, reliability, and cost. Deploy and benchmark inference engines, build compression and quantization workflows, implement acceleration techniques, diagnose serving bottlenecks, and collaborate with kernel, platform, research, product, and customer teams.

Own optimization of LLM and VLM inference systems from model artifacts through production deployment. Improve latency, throughput, memory efficiency, GPU utilization, quality, reliability, and cost per token. Deploy and extend inference engines, build compression and quantization workflows, implement decoding and KV-cache optimizations, create reproducible benchmark harnesses, diagnose production regressions, and collaborate with kernel, platform, research, infrastructure, product, and customer teams.

Design, build, and own reliable production data pipelines for analytics and machine learning using Python and SQL. Implement idempotent data processing, transformations, validation, quality checks, and performance optimization. Orchestrate workloads with Airflow or equivalent frameworks, and run them using Docker and Kubernetes. Collaborate with Analytics, Data Science, and ML teams while contributing technical leadership, CI/CD automation, cloud operations, and cost-efficient infrastructure management.

Investigate complex datacenter incidents involving servers, GPUs, firmware, BIOS/BMC, Linux, and hardware-software interactions. Lead root cause analysis, identify recurring patterns, drive escalations with R&D and ODM vendors, validate firmware rollouts, and create durable fixes. Develop runbooks and troubleshooting guides to improve L1/L2 capabilities. Provide hands-on support during critical incidents and travel to datacenters as needed.

Investigate complex datacenter incidents involving servers, GPUs, firmware, hardware, and Linux. Lead root-cause analysis, identify recurring issues, coordinate evidence-based escalations with R&D and vendors, validate BIOS/BMC firmware rollouts, and create runbooks that improve L1/L2 support. The role includes hands-on datacenter troubleshooting, platform readiness, incident leadership, and technical enablement across Europe and the US.

Own the L3 knowledge system for datacenter servers, GPU platforms, firmware, out-of-band management, and Linux diagnostics. Develop and maintain SOPs, runbooks, troubleshooting guides, and error catalogs; capture findings from L3 investigations; validate procedures with support teams; and prepare documentation for new hardware platforms. The role requires hands-on technical expertise, strong Confluence skills, precise technical writing, and occasional datacenter travel.