Data Scientist 3

About MX 

MX is a leader in open finance, helping banks, credit unions, and fintechs turn raw financial data into clean, connected, and intelligent money experiences. Our data platform powers how millions of people see, understand, and act on their finances. As a Senior Data Scientist on our Chennai team, you’ll work with one of the most comprehensive financial data assets in the industry to build the models and insights that make those experiences smarter. 

The role 

A core data science role: rigorous statistics, machine learning, experimentation, and forecasting applied to real financial-behavior problems. You’ll own problems end to end — from framing and modeling through deployment and monitoring — using generative AI, including agentic workflows, as one tool in the kit. You’ll partner with Analytics Engineering, MLOps, and Product to turn models into measurable impact. 

What you’ll do 

• MX · Build predictive, statistical, and time-series forecasting models of consumer financial behavior  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 → 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 bring 

Required 

• Master’s (Ph.D. a plus) 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 (preferred), 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. 

Preferred 

• 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.