Senior Data Scientist

We are looking for a Data Scientist with hands-on Python coding skills to develop and deploy AI/ML solutions for identity management. The role involves working with real-world identity data, developing NLP and name-matching algorithms, building ML models, and integrating them into production applications. The candidate will also be required to understand, analyze, enhance, and extend an existing Python codebase. Since access to external LLM-based coding assistants such as ChatGPT or Claude may be limited due to the confidentiality of data, the candidate must have strong independent coding and problem-solving abilities.

Responsibilities:

• Develop algorithms for name matching, entity resolution, record linkage, duplicate detection, and identity matching.
• Apply NLP and text similarity techniques, including Jaro-Winkler, Levenshtein, TF-IDF, n-grams, token-based similarity, phonetic matching, embeddings, and transformer-based approaches.
• Perform data preparation, feature engineering, statistical analysis, model training, validation, and performance optimization.
• Develop hybrid solutions combining deterministic rules, fuzzy matching, statistical methods, and machine learning.
• Write clean, efficient, maintainable Python code and develop reusable data science and ML components.
• Understand and work extensively with an existing Python codebase, identify issues, optimize algorithms, and implement new functionality.
• Develop and deploy ML models as production services/APIs and work with engineering teams to integrate them into enterprise applications.
• Analyze model performance and address issues such as false positives, false negatives, scalability, latency, and model explainability.
• Research and evaluate new AI/NLP techniques relevant to identity verification, fraud detection, and identity intelligence.

Requirements:

• Strong hands-on Python programming and software development skills.
• Natural language processing and advanced text matching.
• Strong foundation in data science, machine learning, statistics, and predictive modelling.
• Strong knowledge of NLP, text processing, similarity algorithms, and entity matching.
• Experience with Pandas, NumPy, Scikit-learn, SciPy, and related Python ML/NLP libraries.
• Identity management, digital identity, KYC/e-KYC, or biometric applications.
• Ability to independently understand and modify an existing codebase and develop solutions with limited reliance on LLM coding assistants.
• Experience in fraud detection and anomaly detection models with respect to identity fraud is an added advantage.
• Worked on customer/entity resolution and duplicate identity detection.
• Worked on multilingual NLP, transliteration, and matching of names across different languages/scripts.
• Experience in ML model deployment, REST APIs, Docker, AWS, or MLOps.
• Bachelor's or master's degree in computer science, data science, artificial intelligence, statistics, mathematics, engineering, or a related discipline.

Skills

ML, NLP, Natural Language Processing, Python, Scikit-learn, entity matching, machine learning, predictive modelling, text matching