About the role
Pragmatike is recruiting a Principal ML Ops Engineer for a distributed cloud infrastructure startup building AI-focused services. The role centers on production-grade model serving and infrastructure for real-time AI applications, working with infrastructure, platform, and applied AI teams. The position is listed as fully remote in the EMEA timezone, and Italy appears among the ATS locations.
The engineer will build and operate inference platforms, deployment pipelines, scaling and routing systems, and observability for performance, cost, and reliability. The role also covers GPU optimization, model registries, CI/CD, and the full operational lifecycle, including on-call support. Candidates should bring at least four years in ML Ops, platform engineering, SRE, or related ML infrastructure work, with hands-on model serving and production GPU experience.
Highlights
- Build scalable serving systems using tools such as vLLM, TGI, or Triton.
- Develop canary and blue/green model deployment pipelines.
- Optimize GPU utilization, inference performance, and operating costs.
- Own production ML systems and help set platform engineering practices.
This recap is dataskew's editorial summary, not the company's copy.