Assistant Vice President - Lead Data Scientist - Gurugram

Your role will focus on experimenting with AI Agents and agentic AI, exploring how these technologies can enhance insurance workflows. Rapidly prototype and validate your ideas, demonstrating feasibility and potential impact. Work with large, complex data sets to develop models that solve real business challenges. You will collaborate closely with cross-functional teamsranging from business analysts to senior stakeholdersto translate organizational needs into technical objectives. While the core of your role is research-oriented, collaboration with production teams is essential. You will provide them with tested, validated concepts and support the initial stages of production integration. Partner with software developers, product managers, and other data scientists to ensure AI solutions are aligned with business needs Leverage state-of-the-art AI tools, frameworks, and libraries to accelerate AI development. Document AI research outcomes, development processes, and performance metrics. Present findings to stakeholders in an easily understandable manner

Data Preparation and Feature Engineering

• Conduct data exploration, handling outliers and missing data, to ensure optimal model performance.
• Implement feature engineering techniques and data visualization methods for improved insights.

Predictive and Statistical Modeling

• Perform statistical analysis (t-tests, ANOVA, Regression etc.) to derive actionable insights for internal stakeholders and clients.
• Develop sophisticated predictive models and statistical algorithms for core insurance functions such as persistency, surrender, risk assessment, mortality prediction, and claims optimization.
• Use predictive analytics across pricing, underwriting, claims management, and fraud detection to drive better decision-making.

Machine Learning and Deep Learning Expertise

• Utilize ML and deep learning libraries (NumPy, Pandas, TensorFlow, PyTorch, Scikit-learn) to build, train, and deploy robust models.
• Continually evaluate emerging techniques and frameworks to maintain cutting-edge AI solutions.
• Ensure best practices for model optimization, performance tuning, and deployment efficiency
• Good knowledge on Deep Learning models like ANN, CNN, RNN, Encoders & Decoders and self-attention model like BERT and their transformer family models for text-based analysis and language modeling.

• Programming & Tools: Python, NumPy, Pandas, Scikit-learn, TensorFlow, PyTorch, PySpark
• Generative AI: GPT, VAEs, GANs, diffusion models, fine-tuning pre-trained LLMs (BERT, GPT, T5)
• LLM models APIs (e.g. OpenAI, Anthropic) and LLM Frameworks (e.g. LangChain, LlamaIndex).
• Machine Learning: Regression (linear, logistic), tree-based models (Random Forest, XGBoost), clustering, time series
• Deep Learning: CNN, RNN, LSTM, Transformers, attention mechanisms
• Data Visualization: Matplotlib, Plotly, Dash, Power BI, Tableau
• Databases: SQL (MySQL, PostgreSQL), NoSQL (MongoDB), Vector Databases (Chroma, Weaviate)
• Cloud Services: AWS (S3, EC2), Azure, or GCP (optional based on experience)
• Other: ML Ops (Docker, Git, CI/CD), prompt engineering, RAG (retrieval-augmented generation) architectures