Data Scientist & Analytics Lead- Customer Excellence

Role & responsibilities
A. Propensity Modelling & Predictive Analytics

• Design and develop end-to-end propensity models based recommendation engine and/ or lead scoring engine for products including savings accounts, FDs, personal loans, microfinance, insurance, and gold loans.
• Build binary and multi-class classification models to predict customer likelihood to purchase, activate, or churn.
• Conduct feature engineering using customer demographics, transaction behaviour, bureau data, and digital footprint.
• Regularly validate, recalibrate, and champion-challenger test models to maintain accuracy and lift.
• Deploy models into production pipelines and integrate outputs with CRM, campaign, and operations systems.

B. Lead Scoring Engine

• Architect and implement a dynamic lead scoring framework that ranks prospects and existing customers by conversion probability.
• Integrate lead scoring outputs with the bank's CRM and other core systems for real-time decisioning.
• Define scoring thresholds, segment-wise cut-offs, and feedback loops based on sales team performance data.
• Develop dashboards and reports to track lead funnel performance, score distribution, and model ROI.
• Collaborate with the Sales, Operations, and Technology teams to embed scored leads into automated workflows.

C. Recommendation Engine (Usage & Implementation)

• Conceptualise, design, and implement a hyper-personalized recommendation engine for product suggestions across digital channels (mobile app, net banking, IVR, agent assist).

• Evaluate and recommend appropriate algorithmic approaches collaborative filtering, content-based filtering, hybrid models, or deep learning-based techniques aligned to data availability and business objectives.

• Drive end-to-end implementation of the recommendation engine in collaboration with IT, Digital, and Product teams.

• Define usage guidelines, guardrails, and governance policies for responsible deployment of recommendations.

• Measure recommendation engine performance via CTR, conversion lift, next-best-action acceptance rates, and revenue attribution.

• Continuously improve recommendation relevance by incorporating real-time behavioural signals, seasonality, and campaign feedback.

D. ML based models Workflow Automation

• Develop analytical models to support operational efficiency queue prediction, TAT optimisation, fraud propensity, early warning signals, and NPA prediction.

• Partner with cross-functional teams to identify automation opportunities and build supporting data models.

• Build and maintain data pipelines, feature stores, and model monitoring frameworks.

• Generate insights from structured and unstructured data (call centre logs, app events, branch transactions) to inform process automation priorities.

E. Stakeholder Collaboration & Business Translation

• Work closely with Product, CRM, and business teams to define analytical requirements and deliver actionable insights.

• Present model outputs, recommendations, and business impact to senior management in a clear, non-technical manner.

• Create and manage model documentation, business sign-off artefacts, and model governance records.

Preferred candidate profile

Core Data Science & Machine Learning

Data Engineering & Platforms

Banking Domain Knowledge

1. B.Tech / M.Tech / M.Sc in Computer Science, Statistics, Mathematics, Data Science, or a related quantitative discipline from a reputed institution.

2. Certifications in ML / AI (AWS ML Specialty, Google Professional ML Engineer, Databricks, etc.) will be an advantage.

3. Certifications in Data Science / ML / Cloud (AWS / Azure / GCP) are advantageous.

4. 5 - 10 years of progressive experience in data science, machine learning, and analytics in BFSI (banking, NBFC, fintech, insurance).
   Demonstrated track record of building and deploying propensity or scoring models in a banking or NBFC environment.

5. Prior experience implementing or contributing to a recommendation engine in a digital banking or fintech context will be highly preferred.

Email- [email protected]