Required DevOps - AI Factory
The role:
You will join a global DevOps & Platform team responsible for designing and operating distributed AI production systems at very high scale, with a relentless focus on quality, robustness, and reliability. We provide us with the tools it needs to push forward product innovation and scale out the age of AI. You will own projects end-to-end, drive new implementations forward, and act as a technical partner across the organization. Youll work closely with the Platform teams,AI Labs, Development teams, Architects, AI/ML Researchers, and Product, translating complex needs into resilient systems that scale. Were looking for someone who takes ownership, who can see a problem or a need, shape the solution, and carry it all the way to production.
The day-to-day:
Lead initiatives end-to-end, from design through implementation to production rollout, owning the outcome, not just the task
Design, build, and operate highly available, distributed systems at scale on AWS and GCP, with a strong emphasis on quality and robustness
Work across the full platform stack, Kubernetes, Docker, and Helm, CI/CD and automation, and the data, caching, and messaging layers
Modernize and consolidate different stacks and architectures into our single, coherent platform
Build and operate the infrastructure behind our AI workloads, including GPU inference and the platforms that serve them
Drive system reliability, monitoring, and observability, and troubleshoot complex issues across application, infrastructure, networking, OS, and data layers
Collaborate across Platform, Development, Architects, AI/ML Researchers, and Product, and leverage modern AI coding tools to work at high velocity and quality.
Requirements: 2+ years of hands-on experience in DevOps / Platform / SRE roles, operating distributed production systems at large scale, with a proven ability to lead projects independently and drive new implementations end-to-end
2+ years of experience working with cloud Inferencing services and NeoClouds
2+ years of hands-on experience building and maintaining AI Factory systems on KubeFlow/Ray/similar.
2+ years of hands on experience with GPU Inferencing (vLLM/SGLang/Nvidia Triton)
Deep hands-on expertise with multicloud AWS/GCP/NeoClouds (RunPod, TogetherAI, vast.ai) including Infrastructure as Code (Terraform or similar)
Deep hands-on expertise with Kubernetes and Helm, including a strong understanding of how Kubernetes internals work and how to operate them in production, alongside Docker / containerized systems
Strong knowledge of networking and operating systems (TCP/IP, DNS, load balancing, Linux internals)
Strong experience with CI/CD and automation (GitHub Actions), monitoring and observability (metrics, tracing, logging, tools like Prometheus / Grafana)
High proficiency with current AI coding tools, including a deep understanding of how they work and how to apply them to drive real productivity.
These would also be nice:
Familiarity with AI evals and AI observability system, such as Opik, OpenEval, LangSmith etc
This position is open to all candidates.