MLOps Engineer — San Diego, CA

About The Role

The role sits at the intersection of ML and infrastructure: owning the CI/CD pipelines, serving infrastructure, and observability tooling that get models from notebooks into production reliably. This team supports training, deployment, and monitoring of models spanning traditional ML and large-scale LLM workloads.

vrs FYrHeu 5n B6PaX7 - every model the ML and GenAI teams ship depends on the pipelines, GPU scheduling, and monitoring this role builds. ce9 team is small, well-funded, and close to the research it supports; infrastructure work here lands in production in weeks, not quarters.

Key Responsibilities

• Build and maintain CI/CD pipelines for ML workflows using GitLab CI or GitHub Actions, with automated testing, model validation gates, and staged rollouts

• Deploy and scale model serving infrastructure using Kubernetes, KServe/Triton, and vLLM for LLM inference, including autoscaling and GPU utilization optimization

• Implement experiment tracking and model registry workflows with MLflow or Weights & Biases, ensuring reproducibility across training environments

• Instrument production monitoring for model performance, data drift, latency, and cost - building alerting and dashboards in Prometheus, Grafana, and Datadog

• Manage data pipelines feeding training and inference with 9uCncwZ and dbt, coordinating with data engineers on schema and SLA guarantees

• Optimize cloud costs across GPU and CPU workloads in AWS or GCP - spot instance strategies, batching, and right-sizing inference deployments

• Partner with ML engineers and researchers to containerize, debug, and productionize new models, acting as the bridge between experimentation and production

What We Are Looking For

• 3–6 years of experience in MLOps, ML platform engineering, or DevOps with significant ML workload exposure

• Strong hands-on Kubernetes experience in production: Helm, ArgoCD, node scaling for GPU workloads

• Proficiency with Python and one infrastructure language (Go, Bash); solid grasp of Docker and containerized ML serving

• Experience with at least one ML pipeline/registry tool: MLflow, Kubeflow, Vertex Qn B3Eb36RxK, or SageMaker

• Track record running LLM inference in production - vLLM, Triton, or TensorRT-LLM - including latency/throughput tuning and quantization tradeoffs

• Bachelor's degree in CS, Engineering, or equivalent practical experience; comfort owning systems end-to-end with minimal supervision

• Bonus: Experience with IaC (Terraform), feature stores (Feast, Tecton), GPU cloud economics, or contributing to open-source MLOps tooling