Lead ML Engineer

GOOD TO HAVE

• Understanding of Data Science models
• Exposure to Deep Learning frameworks such as Tensor Flow or Py Torch
• Solid understanding of feature engineering, model evaluation, and experimentation.

• Strong communication and storytelling skills with data
• Ability to work in a collaborative and fast-paced environment
• Passion for solving complex business problems using data
• Lead the design and implementation of production ML pipelines for training, batch inference, and real-time/near-real-time scoring.
• Translate Data Science prototypes into robust, maintainable services and workflows with strong testing, observability, and reliability.
• Build and manage feature engineering workflows, feature stores (where applicable), and reusable ML components.
• Drive model packaging and deployment patterns (containers, serverless, managed endpoints) and optimize for performance and cost.

• Implement CI/CD for ML (model versioning, automated testing, promotion gates, rollback strategies) using Azure Dev Ops / Git Hub Actions integrated with Databricks
• Leverage MLflow (Databricks native) for experiment tracking, model registry, and lifecycle management
• Establish best practices for model monitoring: data drift, concept drift, model degradation, and alerting.
• Define and enforce guardrails for responsible AI: bias checks, explain ability, privacy controls, and auditability.
• Partner with Data Engineering on data quality, lineage, and availability to ensure reliable model inputs.
• Work with Cloud/Platform teams to ensure scalable infrastructure (compute, networking, IAM, secrets, logging).
• Influence target architecture and technology decisions for the ML platform roadmap.
• Provide technical leadership and mentorship to ML Engineers and junior team members.
• Conduct design reviews, code reviews, and establish engineering standards.
• Coordinate delivery plans, estimate work, and manage technical risks and dependencies.

• Languages:
  
  Python (required); SQL; optional Java/Scala
• Experience with MLOps practices, including model versioning, monitoring, and CI/CD for ML pipelines.