Data Scientist with 1+ years in financial analytics, machine learning, and business operations.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Economics graduate from IIT Kanpur (Minor:ML & Management Sciences) with strong foundations in financial analytics, credit risk modeling, and business operations. Proficient in Python, SQL, Power BI, quantitative modeling, and statistical inference with hands-on experience in collections analytics, loan default prediction across real-world finance workflows.
Indian Institute of Technology Kanpur
BS · Economic Sciences
August 1, 2022 – June 30, 2026
Department of Industries, Government of NCT Delhi
AI Research Intern
June 1, 2025 – July 1, 2025
New Delhi, Delhi, India
Maritime Research Center | Think Tank
ML & Economics Research Intern
June 1, 2025 – August 1, 2025
India
Customer Acquisition & Campaign Optimization Platform
June 1, 2026 – July 1, 2026
Processed 45,000+ customer records using PySpark & Spark SQL, automating campaign analytics and reducing effort by 70%. Built XGBoost propensity models identifying the top 20% high-conversion prospects, improving targeting precision by 30%. Developed an A/B testing framework using hypothesis testing and confidence intervals, quantifying a 12% conversion uplift. Designed a Power BI dashboard tracking ROI, conversion, and acquisition KPIs, reducing reporting turnaround time by 80%.
Loan Default Prediction System
March 1, 2026 – April 1, 2026
Built credit risk pipeline estimating Probability of Default (PD) and LGD via LightGBM on 50,000+ commercial lending records. Engineered 15+ features including EMI overdue ratio and charge-off indicators, segmenting top 20% high-default-risk borrowers. Deployed SHAP explainability for borrower risk scoring, reducing manual loan servicing review by 40%.
View ProjectFood Delivery Analytics
January 1, 2026 – February 1, 2026
Built retention pipeline on 7-table schema using cohort analysis and RFM segmentation via NTILE() across 15,000+ records. Engineered churn detection and activation funnel using LAG() & anti-join, identifying at-risk segments across monthly cohorts. Quantified campaign ROI via conditional aggregation, surfacing 2× revenue gap between organic and paid channels. Delivered KPI dashboard covering MoM growth, CLV, and SLA breach rates using CTEs and window functions.
Personalized Offer Ranking
June 1, 2025 – July 1, 2025
Processed 30M+ clickstream and transaction records via Parquet, engineering scalable user/session-level ranking features. Trained LambdaMART with SMOTE-balanced data, optimizing ranking via Optuna-tuned Bayesian hyperparameter search. Validated with 5-fold cross-validation, achieving significant MAP@7 lift over keyword baselines across diverse user cohorts.
View ProjectCultural Fit Analysis
The candidate's project portfolio showcases a strong alignment with data-driven decision-making and analytical roles, which is a good indicator for cultural fit in a data-centric organization. The experience in both academic research (Maritime Research Center) and government (Department of Industries) indicates a breadth of exposure to different organizational structures and problem types. The focus on quantifiable impact in projects suggests a results-oriented approach. The candidate's education from a top-tier institution (IIT Kanpur) and achievements in competitive exams also point to a strong work ethic and intellectual curiosity. However, the experience level is relatively junior, which might require mentorship in a senior role.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through various project descriptions, tackling real-world business challenges with data-driven approaches. Their ability to quantify impact (e.g., 60% reduction in query resolution time, 30% improvement in targeting precision) indicates a results-oriented mindset. The project diversity suggests adaptability and a willingness to learn new domains. However, without direct interview data, assessing collaboration, stress handling, and communication style in a team setting is not possible.