Role overview
We are looking for an ML Engineer with strong MLOps capabilities to own the end‑to‑end machine learning lifecyclefrom data ingestion and feature engineering to deployment, monitoring, and automated retraining. This role requires hands‑on experience building production-grade ML pipelines, not just notebooks or POCs.
You will work closely with data engineering, platform, and business teams to operationalize ML models at scale using the Databricks ecosystem.
What you'll work on
- End-to-End ML Lifecycle Ownership
- Build and maintain pipelines covering data ingestion → feature engineering → model training → deployment → monitoring → retraining.
- Own production ML workflows, ensuring reliability, scalability, and observability.
- Feature Engineering & Feature Store
- Design reusable, production-grade feature pipelines using PySpark.
- Manage feature stores (Databricks Feature Store preferred).
- Ensure point‑in‑time correctness, feature versioning, and online/offline feature parity.
- Model Training & Experimentation
- Develop structured training pipelines (beyond ad‑hoc notebooks).
- Use MLflow for experiment tracking, reproducibility, and lineage.
- Implement hyperparameter tuning and model comparison frameworks.
- Model Deployment
- Build and maintain batch scoring pipelines and real-time inference endpoints.
- Implement CI/CD for ML models using GitHub/Bitbucket.
- Manage model versioning, environment consistency, and deployment automation.
- Model Monitoring & Observability
- Set up monitoring for data drift, concept drift, and model performance degradation.
- Build dashboards and alerting mechanisms for proactive issue detection.
- Automated Retraining
- Design automated retraining pipelines with validation gates and approval workflows.
- Implement trigger-based retraining (data changes, drift thresholds, performance drops).
- Integration with Data Platform
- Work deeply with Databricks (Unity Catalog, Delta Lake, Workflows).
- Understand data dependencies, schema evolution, and platform governance.
- Governance & Compliance
- Ensure model lineage, auditability, and reproducibility.
- Implement access control and governance using Unity Catalog.
What we're looking for
- Real-time inference architecture (REST endpoints, streaming)
- Exposure to GenAI / LLMOps
- Experience with large-scale enterprise ML systems
- Knowledge of cloud platforms (Azure preferred)
Tags & focus areas
Used for matching and alerts on DevFound Fulltime Machine Learning Mlops Data Engineer Ai