About Hivenet
We're building the future of cloud infrastructure. Hivenet is a distributed cloud platform offering GPU computing for AI/ML workloads, secure file storage, and fast file transfer—all powered by a sustainable, decentralized architecture that reduces carbon footprint by 77% compared to traditional data centers. We're a well-funded, remote-first team scaling fast across EMEA.
About the Team
The AI Services team builds Hivenet's AI-native products: managed inference, LLM-as-a-Service, RAG for enterprise, and custom B2B model deployments. We're creating the infrastructure that lets customers run production AI without managing GPUs themselves.
What You'll Do
- Build and operate model serving infrastructure using vLLM, TGI, Triton, or similar
- Design deployment pipelines with blue/green and canary strategies for model rollouts
- Implement auto-scaling, multi-model serving, and intelligent request routing
- Optimize GPU memory management, network throughput, and storage for model artifacts
- Build observability for inference latency, throughput, GPU utilization, and cost attribution
- Manage model registries and CI/CD pipelines for model deployments using KubeAI, Kubeflow, or MLflow
What You Bring
- 3+ years of experience in ML Ops, Platform Engineering, or SRE for ML systems
- Hands-on experience with model serving frameworks (vLLM, TGI, Triton, or similar)
- Strong container orchestration skills and experience operating GPU workloads at scale
- Experience with MLOps tooling: model registries, experiment tracking, deployment automation
- Proficiency in Python and infrastructure-as-code tools
- Demonstrated ability to use AI coding assistants effectively—rapidly prototyping, debugging, and refactoring while maintaining strong engineering judgment and production standards
Bonus Points
- Experience with KubeAI, Kubeflow, MLflow, or similar ML platforms
- Background in GPU scheduling, memory optimization, or multi-tenant inference
- Familiarity with cost optimization across GPU types (CUDA/ROCm environments)
The Reality
- This is an early-stage startup—you'll wear many hats
- We move fast and expect ownership mentality
- You'll be building processes, not following existing ones
- You'll own the full lifecycle: build it, ship it, run it, fix it at 2am if needed
What We Offer
- Full remote work from anywhere (EMEA timezone preferred)
- Equipment budget to set up your ideal workspace
- Company offsites to connect with the team
- Career growth in a scaling engineering organization
- Work on cutting-edge distributed systems and AI infrastructure
