N
Nebius Tech Stack
Full-stack AI cloud infrastructure for model training and deployment
Nebius operates a GPU-backed AI cloud platform targeting ML practitioners at startups, enterprises, and research institutions. The tech stack—PyTorch, Ray, Slurm, MLflow, NVIDIA H200 GPUs, and Kubernetes—is purpose-built for distributed training and inference workloads. With 96 engineers and 30 ops staff actively hiring across 17 countries and projects spanning hardware validation, data center buildout, and regional go-to-market expansion, Nebius is scaling toward infrastructure-as-a-service for generative AI at scale.
Tech Stack 158 technologies
Core StackAzure AD Python HubSpot Linux MLflow PyTorch Kubernetes Go Azure Functions Terraform C# Microsoft Sentinel Microsoft Defender XDR Microsoft Purview Kusto Query Language PowerShell Microsoft Graph API Bash Azure Logic Apps QEMU KVM Ray Slurm NVIDIA SQL H200 GPU API Management Azure Service Bus Event Grid REST API+122 more
AdoptingNVIDIA
What Nebius Is Building
◆Challenges
- Reducing infrastructure costs
- Scaling inference platform usage across regions
- Improving sla performance
- Building l3 support line from scratch
- Fleet stability
- Reducing resolution times
- Hardware firmware issues
- Large-scale ai deployments
- Optimizing gpu performance for ml training
- Building large in-house ai/ml teams
▲Active Projects
- Us market pipeline development
- Hardware validation rollout
- Ats to hris workflow ownership
- Design partner initiatives and joint campaigns
- Hardware reliability strategy
- Post-hire quality survey design
- Low precision training & inference
- Ai data center development
- Implementing asset management processes
- Regional go-to-market strategy
Hiring Activity
220 roles · 140 in 30d
Department
Seniority
Notable leadership hires: Warehouse Operations Lead, Regional Sales Director, Director GTM M&A
Company intelligence
