Key Responsibilities
- Build and optimize ranking, recommendation, personalization, and search relevance models.
- Develop feature pipelines using behavioral, transactional, and content data with strong data-quality and leakage controls.
- Develop classical ML models using XGBoost, LightGBM, linear/logistic models, and Learning-to-Rank.
- Build retrieval, ranking, and re-ranking systems using BM25, embeddings, ANN, and hybrid retrieval.
- Perform offline evaluation using NDCG, MAP, Recall@K, calibration, and slice analysis.
- Design and analyze A/B tests and online experiments.
- Deploy and monitor models through real-time APIs and batch pipelines.
- Implement MLOps, including model versioning, monitoring, experiment tracking, and retraining.
- Partner with Product and Engineering teams to drive measurable business impact.
Mandatory Qualifications
- 10+ years of Machine Learning experience with production ML systems.
- Strong production experience in recommendation, ranking, personalization, or search.
- Strong Python and SQL skills.
- Expertise in scikit-learn, pandas, XGBoost/LightGBM, and working knowledge of PyTorch/TensorFlow.
- Strong understanding of information retrieval, Learning-to-Rank, and ranking metrics.
- Hands-on A/B testing and experimentation experience.
- Experience with Spark/distributed data processing.
- Experience deploying ML models on AWS/SageMaker or equivalent cloud platforms.
- Strong MLOps, statistics, probability, and Bayesian methods knowledge.
- Ability to clearly communicate ML concepts and trade-offs to technical and business stakeholders.