The individual is responsible for developing, implementing and maintaining knowledge-based or artificial intelligence application systems.
Required Qualification
3 years (Mid) / 5 years (Senior) experience shipping ML systems into production (not just notebooks).
Strong programming skills in Python and familiarity with ML libraries (e.g., scikit-learn, PyTorch, TensorFlow, XGBoost).
Experience with data processing and analytics using tools such as Pandas, NumPy, SQL, Spark (as applicable).
Solid understanding of ML fundamentals: bias/variance, evaluation metrics, cross-validation, feature engineering, and error analysis.
Experience building and deploying ML services (e.g., FastAPI/Flask), containerization (Docker), and CI/CD.
Ability to communicate tradeoffs and results clearly to both technical and non-technical stakeholders.
Key Responsibilities
Model Development & Applied AI
Partner with stakeholders to frame business problems as ML/AI use cases, define success metrics, and identify required data.
Build and iterate on models using appropriate approaches (e.g., classification, regression, ranking, clustering, anomaly detection, NLP).
Perform feature engineering, dataset creation, labeling strategies, and model evaluation with strong scientific rigor.
Implement techniques for model interpretability, bias assessment, and responsible AI where applicable.
Production Engineering & MLOps
Build production-grade ML services and pipelines (batch real-time), ensuring performance, reliability, and maintainability.
Deploy models to cloud environments using CI/CD and infrastructure-as-code best practices.
Implement monitoring for data drift, model drift, latency, throughput, cost, and model quality.
Maintain versioning for datasets, features, models, and experiments to ensure repeatability and governance.
Data & Platform Collaboration
Collaborate with data engineering to create robust data pipelines and ensure data quality.
Work with software engineers to integrate ML into applications through APIs, event streams, or workflow orchestration.
Document architecture, operational runbooks, and model cards; participate in reviews and knowledge sharing.