Senior Infrastructure/Platform Engineer

• Experience running large-scale systems in production with a real reliability practice: error budgets, runbooks that hold up under incident pressure, data-backed capacity planning
• Background at smaller, fast-moving companies, knows when “shipped” beats “polished” and raises slipping timelines early
• Comfortable working in an agent-heavy workflow: delegating substantial work to AI agents, critically reviewing their output, and recognizing current limitations
• Strong skills in container orchestration (Kubernetes)
• Strong skills in infrastructure-as-code (Terraform or similar)
• Strong GPU infrastructure experience
• Proficiency in at least one lower-level systems programming language (Rust preferred; Go, C++, or Zig acceptable)
• Multi-cloud experience is a plus
• Open-source contribution history is a plus

In return we offer:

• The friendliest community of like-minded IT-people
• Open knowledge-sharing environment - exclusive access to a rich pool of colleagues willing to share their endless insights into the broadest variety of modern technologies
• Mobilunity Medical Insurance program designed to attend our teams’ needs
• Paid vacations and sick leaves, including 5 paid days per year that don’t require a sick note
• Perfect office location in the city-center (900m from Lukyanivska metro station with a green and spacious neighborhood) or remote mode engagement: you can choose a convenient one for you, with a possibility to fit together both
• No open-spaces setup - separate rooms for every team’s comfort and multiple lounge and gaming zones
• English classes in 1-to-1 & group modes with elements of gamification
• Neverending fun: sports events, tournaments, music band, multiple affinity groups

On behalf of our Client from USA, Mobilunity is looking for a Senior Infrastructure/Platform Engineer

Our client builds a self-hostable inference engine for pre-trained models, a wide catalog of embedding, reranking, extraction, and OCR models packed onto shared GPUs at high utilization, letting customers run them in their own cloud at a fraction of the cost of managed APIs. Search and document processing is the initial focus area, though the engine is general-purpose. The project is open source under a permissive license, holds a recognized security compliance certification, and has an actively growing developer community. The company is based in the US and has raised a sizable round from well-known venture investors. You’ll help build the engine and own how it gets deployed and operated. That spans the deployment tooling shipped to customers (infrastructure-as-code modules, Kubernetes packaging, deployment guides, air-gapped install paths) and the internal platform behind it (build/release pipelines, multi-architecture GPU images, model caching, evaluation infrastructure). Deployments currently target two major public clouds, with additional providers, including specialized GPU cloud providers.

As the company moves toward offering managed clusters, you’ll help define the operational bar that makes that credible: service-level objectives, on-call practices, scale-from-zero behavior, observability, and GPU cost efficiency.

The team works in short, multi-week milestones with high autonomy and a workflow built heavily around AI coding agents (substantial token usage weekly). Senior engineers own requirements end-to-end across multiple projects.