DEVOPS ENGINEER (AI, ML)

Mission:

You will play a key role in implementing advanced AI solutions, fine-tuning large language models, and crafting customer-facing demos that showcase the power of our hardware.

What You Will Do:

• Experiment & bundle: Run systematic experiments with LLM/RAG software stacks on Comino servers & workstations; benchmark throughput, latency, power draw, thermals and assemble verified AI bundles (hardware + pre-configured software).

• Build and maintain production services on this hardware (inference APIs, vector-DB back-ends, CI/CD agents).

• Automate AI/MLOps pipelines (MLflow, BentoML, KServe); deliver end-to-end tests via Docker, K8s, GitHub Actions.

• Execute training, fine-tuning, distillation, quantization and vector-based RAG workflows.

• Operate infrastructure: Terraform, K8s, Helm, Istio; GPU/CPU autoscaling, cost & drift monitoring.

• Passion for AI and a deep interest in learning and developing in the area. You are the kind of person who gets excited about the possibilities AI offers.

• DevOps/MLOps experience, including 1+ year in production ML.

• Deep interest in AI and hands-on experience training or fine-tuning models.

• Write code and infrastructure with AI assistants (Copilot, GPT-4o, etc.).

• Proven experience deploying local LLMs and building AI services.

• Familiarity with GPU cluster orchestration, containerization, and Linux systems.

• Strong experience with Docker, Terraform, Helm, Prometheus/Grafana.

• Personal projects on inference distillation/quantization.

• Nice to have: Hands-on with latest GPUs (H100, MI300).