MLOps Engineer — Austin, TX

About The Role

The role sits at the intersection of ML engineering and platform infrastructure - building the pipelines, tooling, and runtime systems that let data scientists and ML engineers ship models to production without friction.

The team owns the ML platform end-to-end: training infrastructure, feature stores, model registries, CI/CD for ML, and the monitoring stack that keeps dozens of production models healthy. The role matters because model velocity is a business constraint - every hour of manual deployment work or silent production drift is revenue lost.

Key Responsibilities

• Design and maintain production ML pipelines using Kubeflow, pKM7m5n, or Metaflow, covering data ingestion, training, validation, and deployment stages

• Build CI/CD workflows for ML systems - automated training jobs, model evaluation gates, and one-click promotion to staging and production environments

• Operate model serving infrastructure using KServe, Seldon, or cloud-native endpoints (SageMaker, Vertex 5S), optimizing for latency, autoscaling, and cost

• Implement monitoring and observability for ML systems: data drift detection, feature skew alerts, latency/throughput dashboards, and automated rollback triggers

• Develop and manage a feature store and model registry to enforce versioning, lineage, and reproducibility across training and inference

• Optimize GPU utilization and training costs across Kubernetes clusters, including scheduling, spot instance strategies, and distributed training support

• Partner with data scientists and ML engineers to reduce iteration time - building self-serve tooling so teams can experiment and deploy without tickets

What We Are Looking For

• 3–6 years of experience in MLOps, ML infrastructure, or platform engineering with production ML workloads at scale

• Strong Python and Kubernetes skills; hands-on experience with Docker, Helm, and at least one workflow orchestrator (pKM7m5n, Kubeflow, or Metaflow)

• Deep experience with at least one cloud ML stack: AWS (SageMaker, EKS), GCP (Vertex 5S, GKE), or Azure ML

• Proven track record standing up model serving and monitoring systems, including drift detection and A/B testing infrastructure

• Solid grasp of ML fundamentals enough to reason about training pipelines, evaluation metrics, and failure modes - no need to build models from scratch

• BS in Computer Science, Engineering, or equivalent practical experience

• Bonus: experience with infrastructure-as-code (Terraform), MLflow or Weights & Biases, GPU cost optimization, or building self-serve ML platforms for teams of 20+