MLOps Engineer

Skills Matrix Skill Last Used Experience in Years/month Rating (10 points) 1 = newbie 10 = expert Hands on Exp. Yes/No LLM GKE or Cloud Run BigQuery / BigQuery ML Job Description: The MLOps Engineer is responsible for operationalizing, scaling, and maintaining enterprise AI/ML systems across cloud, hybrid, and on-premise environments. The role focuses on enabling reliable delivery of LLM workloads, retrieval-augmented generation (RAG), document intelligence, multimodal processing, and predictive/ML pipelines-supported by strong governance, observability, security, and automation. Key Responsibilities: Build and automate end-to-end ML pipelines (data ingestion → feature engineering → training → evaluation → packaging → deployment). Establish model CI/CD workflows including versioning, automated testing, canary/blue-green deployments, and rollback strategies. Operationalize LLM-based and RAG systems (embedding workflows, vector indexing, latency optimization, grounding quality checks). Productionize document-processing and multimodal workflows (OCR parsing, enrichment flows, batch/stream scaling). Implement observability (data quality, drift, safety indicators, inference latency, error conditions). Enforce Responsible AI controls (auditability, reproducibility, governance metadata, lineage, approval workflows). Maintain secure serving environments (container hardening, IAM, secrets, network isolation). Optimize GPU/CPU utilization, autoscaling, throughput, and cost efficiency. Create reusable templates, reference architectures, starter repos, and documentation. Required Skills & Qualifications: Strong Python, CI/CD, Docker, Kubernetes. Experience operationalizing LLM, RAG, and predictive ML systems. Strong foundations in data engineering, schema governance, batch/stream pipelines. Security mindset (PII controls, secrets, network boundaries, auditability). Vertex AI (ML orchestration & CI/CD, training, tuning, deployment, model registry & monitoring). BigQuery / BigQuery ML (analytics & in-warehouse ML). Cloud Composer + Dataflow (batch/stream ETL orchestration). GKE or Cloud Run (secure, scalable model serving). Artifact Registry + Cloud Build/Cloud Deploy (container & CI/CD). Preferred Qualifications: Familiarity with agentic reasoning patterns and workflow chaining. Experience with LLM evaluation, grounding, bias/safety checks. Contributions to open-source ML/MLOps tooling.

Created: 2026-03-04