GenAI MLOps Engineer

Job Description

About the Role As a GenAI MLOps Engineer on our AI Engineering team, youll build, deploy, and maintain the core infrastructure that powers our generative-AI products. Youll partner closely with data scientists and software engineers to productionize LLM-based models, automate workflows, and keep services reliable and cost-effective. Core Responsibilities (Must-Have) Pipeline & CI/CD Design, build, and operate repeatable ML pipelines (data prep training evaluation deploy) using tools such as Airflow, Prefect, or cloud-native solutions. Author automated CI/CD workflows (GitHub Actions, Azure DevOps, or Jenkins) for model code, pipelines, and container builds, including linting and automated tests. Model Deployment & Serving Containerize models with Docker; deploy to Kubernetes (AKS/EKS/GKE) or serverless (Cloud Run, Azure Functions). Implement safe rollout patterns (canary, blue/green) to minimize risk when updating model versions. Monitoring & Alerting Instrument inference endpoints and pipelines with key metrics (latency, throughput) and logs. Create dashboards and alerts (Prometheus/Grafana or cloud-native alternatives) to detect errors, drift, and performance regressions. Cloud & Infrastructure Operate core compute resources on one major cloud platform (Azure Databricks, AWS SageMaker, or GCP Vertex AI). Write and maintain basic Infrastructure-as-Code (Terraform, or CloudFormation) for provisioning clusters and managed services. GenAI Orchestration & Vector Retrieval Use orchestration framework (e.g., LangGraph, Langfuse etc) to automate GenAI workflows. Support embedding-based retrieval pipelines: collaborate on vector index maintenance and refresh processes. Collaboration & Documentation Work with data science to integrate new models into production. Produce clear runbooks, architecture diagrams, and “on-call” guides. Qualifications 5 years in DevOps/MLOps roles, including at least 3 years supporting ML or deep-learning systems. Hands-on with one major cloud (Azure/AWS/GCP) and experience provisioning compute for training/inference. Strong skills in Docker and Kubernetes or serverless deployments. Proven ability to author CI/CD pipelines and IaC. Experience with monitoring stacks (Prometheus/Grafana, Datadog, or cloud-native tools). Familiarity with a prompt-orchestration framework (e.g., LangChain) and core vector-retrieval concepts. Soft Skills Effective communicator who can translate technical details to cross-functional teams. Strong problem-solver who can troubleshoot across data, model, and infrastructure layers. Eager to learn new tools and iterate rapidly in a fast-paced environment. Additional Information Growth & Nice-to-Have Security & Compliance: Basic API auth (OAuth/JWT), secrets management (Key Vault, AWS KMS), and data encryption. Testing & Validation: Data-quality checks (e.g., Great Expectations), adversarial testing, and automated model-quality gates. Scalability & Cost Optimization: Capacity planning, load testing (Locust, JMeter), spot-instance usage, and caching strategies (Redis). Reliability Engineering: Participation in on-call rotations, post-mortems, and error-budget SLOs; chaos-testing fundamentals. Experimentation Lifecycle: Tracking experiments and hyperparameter sweeps (MLflow, Optuna), and supporting A/B tests. Tooling Flexibility: Familiarity with alternative MLOps frameworks (Kubeflow, TFX) or observability stacks (OpenTelemetry). Our Benefits Flexible working environment Volunteer time off LinkedIn Learning Employee-Assistance-Program (EAP)