• What you’ll do


ML Pipelines:
• Build ML pipelines for data ingestion, processing, model deployment, and evaluation
• Own CI/CD for ML systems, including automated testing, model versioning, and deployment workflows
• Implement monitoring for model performance, latency, throughput, and costs with budget alerting
• Set up experiment tracking and model registry systems (MLflow, Weights & Biases, or similar)
• Define and monitor SLIs/SLOs for production model serving


Infrastructure & Orchestration:
• Manage Kubernetes and Slurm clusters for GPU workloads with multi-tenant resource allocation
• Optimize GPU utilization and implement cost controls across training and inference workloads
• Own CI/CD pipelines, model versioning, and deployment automation


Model Serving & APIs:
• Deploy and optimize LLM serving infrastructure using vLLM
• Apply inference optimizations: quantization, continuous batching, PagedAttention, KV cache management to maximize throughput and minimize latency
• Design and build production-grade async API services (FastAPI, etc.) with pre/post-processing, business logic, and strict latency SLAs
• Continuously optimize serving costs through model compression, batching strategies, and infrastructure tuning
• Implement A/B testing infrastructure and canary deployments for safe model rollouts


Enablement & Best Practices:
• Create templates and documentation to accelerate team productivity
• Establish MLOps best practices and guide teams in their adoption
• Support model training experiments when needed 

Minimum qualifications

• 5+ years in MLOps, DevOps, or platform engineering with focus on ML workloads
• Expert-level experience deploying and optimizing LLM serving infrastructure
• Strong Python skills with experience building production APIs (FastAPI or similar)
• Proven experience with cost optimization for GPU-intensive workloads: tracking, budgeting, alerting, and resource efficiency
• Hands-on experience with Kubernetes and Docker for GPU workloads
• Experience with job orchestration systems (Slurm, Ray, Argo, Kubeflow, or similar)
• Solid understanding of monitoring and observability for production ML systems
• Naturally curious with a track record of proactively identifying and implementing improvements

Preferred qualifications

• Deep knowledge of GPU architectures and their performance implications for inference optimization
• Expertise in model compression techniques: quantization (INT8, INT4, FP8), pruning, distillation for production deployment
• Understanding of security best practices for ML serving: authentication, authorization, rate limiting, model access controls
• Experience managing multi-tenant GPU clusters with fair scheduling and resource isolation
• Proficiency with infrastructure-as-code tools 
• Experience supporting distributed training infrastructure: multi-node job orchestration, checkpoint management, debugging training failures
• Contributions to open-source MLOps tools or serving frameworks

What we offer

• Critical infrastructure ownership for high-impact AI projects that are strategically vital to the company, with direct visibility to senior leadership including the CEO
• State-of-the-art GPU infrastructure: H200 fleet, vLLM serving stack, cutting-edge optimization tools
• Expert ML team who have released top Hugging Face models, published at NeurIPS, and built production systems that will run on your infrastructure
• Significant autonomy in designing MLOps solutions, choosing tools, and shaping infrastructure strategy for LLM serving
• Modern tooling: Latest MLOps frameworks, coding assistants, best-in-class development environment
• Hybrid work model with our Amsterdam office - home to the AI House, bringing together 200+ AI professionals through events and collaborations
• Competitive compensation, top-spec MacBook Pro, and an environment genuinely built for professional growth and learning


If you're passionate about building scalable, high-performance infrastructure that enables cutting-edge AI deployment and want to see your work impact millions of users globally, let's talk.