Manager – ML Engineering

Key Deliverables in this role

MLOps & Production ML Pipelines

• Design and implemented-to-end MLOps pipelines covering data ingestion, feature engineering, model training, evaluation, deployment, and monitoring.
• Build and maintain CI/CD pipelines for ML models with automated testing, validation, and rollback capabilities.
• Implement experiment tracking, model versioning, and data versioning using tools such as ML flow, DVC, Weights & Biases, or similar.
• Manage model registry workflows - promoting models from experimental staging production with governance and approval gates.
• Automate reproducible training runs with parameterized configs, seed management, and environment pinning.

Productionising & Scaling ML Models

• Deploy and serve ML models at scale using AWS SageMaker, SageMaker Endpoints, or equivalent managed ML platforms.
• Optimize model inference for latency, throughput, and cost - including quantization, distillation, batching strategies, and GPU/CPU optimization.
• Containerize ML workloads using Docker and orchestrate with Kubernetes for reliable, reproducible deployments.
• Build and maintain RESTful / gRPC model serving APIs with proper error handling, authentication, and rate limiting.

Monitoring, Observability & Reliability

• Set up continuous monitoring for data drift, model drift, and performance degradation in production - critical in FMCG where consumer behaviour and market dynamics shift rapidly.
• Build alerting and automated retraining pipelines triggered by drift detection or performance thresholds.
• Ensure model observability through logging, metrics dashboards, and explainability tooling.
• Define and track ML-specific SLOs/SLAs (latency p50/p95/p99, throughput, accuracy, freshness).

Infrastructure & Tooling

• Maintain and improve internal ML platforms and tooling to accelerate team productivity.
• Drive best practices for reproducible experimentation, code quality, testing, and documentation across the ML team.
• Evaluate and integrate emerging tools and frameworks into the AI stack as the ecosystem evolves.

Critical success factors for the role

• 3+years of hands-on experience in Machine Learning / Deep Learning engineering in production environments.
• Demonstrated track record of taking AI/ML models from prototype to production at scale.
• Experience in FMCG, consumer goods, supply chain, or retail domains is a strong plus.

Desirable success factors for the role

• Exposure to modern data engineering concept and data preparation to create ML-ready data sets
• Familiarity with data governance / metadata practices(catalog, lineage, stewardship).
• Experience with additional AWS services (e.g.,Lambda, Redshift) or Azure/GCP.
• Relevant certifications: AWS / Snowflake / SnapLogic.

Core Technical Skills

• Python : Strong proficiency; clean, production-quality code with solid software engineering practices (OOP, design patterns, testing).
• MLOps Tools: Production experience with MLflow, DVC, Kubeflow, or W&B for experiment tracking, model registry, and versioning.
• Cloud ML platforms: Hands-on with AWS SageMaker(Training, Endpoints, Pipelines, Registry) or equivalent (GCP Vertex AI, Azure ML).
• Model Lifecycle: End-to-end ownership data exploration, experimentation, training, evaluation, deployment, monitoring, and retraining.
• Containerization: Experience with Docker and Kubernetes for packaging and deploying ML workloads.
• CI / CD for ML: Experience buildingCI/CD pipelines (GitHubActions, Jenkins, GitLabCI) integrated with ML workflows.
• Version Control: Proficient with Git; experience with Git-based workflows for code, data, and model artifacts.