Data scientist

Key Responsibilities:

• Build, train, and validate machine learning models for prediction, classification, and clustering to support NBA use cases.
• Conduct exploratory data analysis (EDA) on both structured and unstructured data to extract actionable insights and identify behavioral drivers.
• Design and deploy A/B testing frameworks and build pipelines for model evaluation and continuous monitoring.
• Develop vectorization and embedding pipelines using models like Word2Vec, BERT, to enable semantic understanding and similarity search.
• Implement Retrieval-Augmented Generation (RAG) workflows to enrich recommendations by integrating internal and external knowledge bases.
• Collaborate with cross-functional teams (engineering, product, marketing) to deliver data-driven Next Best Action strategies.
• Present findings and recommendations clearly to technical and non-technical stakeholders.

Required Skills & Experience:

• Strong programming skills in Python, including libraries like pandas, NumPy, and scikit-learn.
• Practical experience with text vectorization and embedding generation (Word2Vec, BERT, SBERT, etc.).
• Proficiency in Prompt Engineering and hands-on experience in building RAG pipelines using LangChain, Haystack, or custom frameworks.
• Familiarity with vector databases (e.g., PostgreSQL with pgvector, FAISS, Pinecone, Weaviate).
• Expertise in Natural Language Processing (NLP) tasks such as NER, text classification, and topic modeling.
• Sound understanding of supervised learning, recommendation systems, and classification algorithms.
• Exposure to cloud platforms (AWS, GCP, Azure) and containerization tools (Docker, Kubernetes) is a plus.

Experience – 5+ years