Mlops Engineer Jobs

About The Position

Requirements

• 5+ years of production Python engineering
• Strong OOP fundamentals: classes, inheritance, composition, design patterns
• Testing discipline: unit, integration, fixtures, mocking
• Demonstrated async and parallel optimization (asyncio, multiprocessing, threading)
• Building and operating production Python services (APIs, workers, background jobs)
• Familiarity with FastAPI or Flask
• Experience deploying to self-hosted/on-prem environments
• Translate engineering needs into clean, maintainable code
• Comfortable coaching peers on production engineering practices
• Curious about ML and motivated to ramp into it

Nice To Haves

• Prior MLOps or ML platform experience
• ML frameworks: scikit-learn, XGBoost, PyTorch
• Observability stack: Prometheus, Grafana, structured logging/tracing
• RAG pipelines: vector stores, semantic search
• LLM serving: vLLM, Text Generation Inference
• GenAI/agentic frameworks: LangChain, LlamaIndex, DSPy
• Orchestration: Prefect, Kubeflow, Airflow, or similar
• Kubernetes and containerization in on-prem environments
• Experiment tracking: MLflow
• LLM observability: Phoenix, Langfuse, OpenLIT
• On-prem GPU infrastructure management

Responsibilities

• Build automated ML pipelines: data ingestion → training → evaluation → deployment → retraining
• Deploy and serve models (batch + real-time) via FastAPI/Flask APIs with auto-scaling and rollback
• Implement CI/CD for ML: model packaging, versioning, automated deployments
• Optimize workflows using async, parallelism, Ray, and Dask
• Design reusable internal Python packages for preprocessing, training, inference, and evaluation
• Refactor data science notebooks into maintainable OOP modules
• Build workflow orchestration for training and inference pipelines
• Create standardized templates for model development
• Monitor latency, drift, data quality, and model performance
• Build alerting for degradation and anomalies (Prometheus, Grafana)
• Create dashboards for production model health
• Set up automated retraining triggers
• Coach data scientists on production-grade Python: testing, OOP, async/parallel patterns
• Establish and enforce software best practices across the ML codebase
• Partner with data scientists to translate pain points into engineering solutions

Benefits

• Comprehensive health benefits
• 401(k) plan with a company match
• Employee stock purchase plan
• Vacation time
• Sick time
• Paid holidays