Data Scientist III

Overview

Senior Data Scientist on the Chennai team building predictive, statistical and forecasting models of consumer financial behavior, owning them end to end and applying generative AI where it adds value.

What you'll do

• Build predictive, statistical, and time-series forecasting models of consumer financial behavior, including transaction intelligence and enrichment, personalization, cash-flow and financial-insight forecasting, and fraud / anomaly detection.
• Own models end to end: problem framing, feature engineering, training, evaluation, deployment, and monitoring, including drift detection (data and concept), alerting, and retraining.
• Design and analyze experiments (A/B, multivariate) and apply causal inference to measure real impact.
• Use generative AI where it adds value (LLMs, RAG, and agentic tool-using, multi-step workflows) to extract signal from unstructured financial text and automate analysis, with sound evaluation and guardrails.
• Build and serve models on Google Cloud (Vertex AI): training, registry, endpoints, and model monitoring, with explainability (SHAP / LIME) and documentation for a regulated industry.
• Communicate findings as clear, actionable stories for peers, product partners, and senior leaders.

What you'll need

• Master's in a quantitative field, or equivalent experience.
• 5+ years applying data science to real business problems, with models that moved meaningful metrics.
• Strong foundations in statistical and causal inference, experimentation, and hypothesis testing.
• Breadth across statistical and machine-learning methods: regression / GLMs, tree ensembles, clustering, and time-series forecasting (e.g., ARIMA, Prophet, gradient-boosted or deep-learning approaches), with sound model evaluation and validation.
• Proficiency in Python (pandas, scikit-learn, PyTorch and/or TensorFlow) and SQL.
• Experience taking models to production and keeping them healthy: deployment, monitoring, and drift capture / retraining, ideally on Google Cloud Platform with Vertex AI; other clouds considered.
• Applied fluency with generative AI (prompting, RAG, LLM evaluation) and hands-on experience with agentic workflows and orchestration frameworks (e.g., LangGraph, CrewAI).
• Clear communication of complex results to technical and non-technical audiences.

Nice to have

• Ph.D. in a quantitative field.
• Fintech / banking domain: credit / risk scoring, fraud / AML, or consumer financial data.
• Vertex AI Pipelines, Feature Store, or Model Monitoring; dbt / Airflow (Cloud Composer).
• Explainable AI and model governance in a regulated setting.
• Multi-agent orchestration, agent evaluation / observability, or MCP-style tool integration.

Details

• Location: Chennai, Tamil Nadu, India.
• Work mode: in office for a standard 40-hour workweek.
• Applicants must have the legal right to work in India without sponsorship; no work sponsorship is provided.