Data Scientist focused on building practical, high-impact analytics and ML solutions.
I work across the full lifecycle of modeling: framing the problem, shaping data, designing robust features, training and validating models, writing scoring code, and deploying monitoring frameworks that survive real-world business use.
My interests pull me toward the frontier—places where the problem isn't fully solved and where thoughtful modeling, experimentation, and system design matter.
Risk Modeling & Credit Analytics
- Scorecards for collections, acquisitions, and behavior.
- Variable selection, documentation, model monitoring, UAT, and production deployment.
- Strong grounding in risk thinking, decision systems, and practical constraints of lending or Anaytics.
Machine Learning & Applied Modeling
- Regression, classification, clustering, .
- LSTM/GRU time-series/sequential modeling for macroeconomic modelling region wise.
- Explainability research (XAI).
Data Engineering & ML Engineering
- SQL, Python, SAS.
- ML pipelines, reproducibility, code hygiene, versioning, containerization basics.
AI/ML for Lending
- Designing frameworks that decide who to lend to, how much, and through which channel.
- Current goal: build a POC solution powered by transparent, data-driven decision engines.
- Credit Scorecard Development – End-to-end model development for collections and acquisition strategies.
- GRU-based Stock Prediction Model – Sequential modeling with explainability add-ons.
- Yoga Learning App Prototype – Gamified learning using computer vision for posture feedback.
- RAG and LLM Experiments – Practical experiments in retrieval, prompting, and evaluation.
- Personal ML Utilities – Feature engineering helpers, monitoring templates, and modeling toolkits.
Languages: Python, SQL, SAS
ML: XGBoost, LightGBM, Random Forests, Scikit-learn, GRU/LSTM
Data: Pandas, NumPy
Other Interests: complexity, causality, economics, philosophy, and systems thinking
Building, learning, and improving—one model, one idea, one experiment at a time.