Data Scientist · ML Engineer · AI Systems Builder
Final-year B.Tech Computer Science student at Manav Rachna University, Faridabad Building production-grade ML systems that turn raw data into measurable business impact.
I design and ship end-to-end machine learning systems — from raw dataset ingestion through model selection, explainability, and containerised API deployment. My work sits at the intersection of applied ML and software engineering: I care equally about model performance and production readiness.
Currently focused on MLOps, LLM applications, and explainable AI. I am actively seeking opportunities in Data Science, ML Engineering, and AI research — both in industry and in graduate programmes.
"Delivered AI systems with 80%+ efficiency gains across 100K+ record datasets."
Production-grade automated ML platform for predicting, explaining, and preventing customer churn.
A full MLOps pipeline built on 7,043 customer records — not a notebook, but a deployable system with:
- 4 model comparison (Logistic Regression, Random Forest, XGBoost, CatBoost) with stratified 5-fold CV + SMOTE
- SHAP explainability per customer prediction with 12 domain-driven retention action recommendations
- MLflow experiment tracking, model versioning, auto-promote, and rollback
- PSI-based data drift detection with three-tier alerting (Stable / Moderate / Significant)
- Flask REST API (7 endpoints, Pydantic v2 validation) + Streamlit dashboard (6 pages)
- 96-test CI/CD pipeline with GitHub Actions → Docker Compose deployment
- 🎯 Target metrics: ROC-AUC ≥ 0.85 · F1 ≥ 0.70
Full-stack University ERP Dashboard with AI-powered features and face recognition.
A production-deployed system built for Manav Rachna University's Computer Science department:
- AI course code generation via Google Gemini with university-standard format enforcement
- Face recognition for student identification using face-api.js
- JWT authentication, Supabase (PostgreSQL), bcrypt security, rate limiting
- Full-stack: Next.js 15 + Express 5 + TypeScript, deployed on Vercel + Render
B.Tech — Computer Science & Engineering Manav Rachna University, Faridabad · 2022 – 2026 (Expected)
Relevant Coursework: Machine Learning, Data Structures & Algorithms, Database Management Systems, Statistics for Data Science, Software Engineering
| Metric | Value |
|---|---|
| 🗂 Public Repositories | 16 |
| 🧪 Tests Written | 96 (Customer Churn project alone) |
| 📦 Production Deployments | 2 live apps (Streamlit Cloud · Vercel) |
| 🤖 ML Models Compared | 4 algorithms per project with cross-validation |
| 📡 API Endpoints Built | 7 (REST) + 13 (University ERP) |
| 🐳 Containerised Projects | Docker + Docker Compose (multi-service) |
| 🔁 CI/CD Pipelines | GitHub Actions (lint → test → Docker build) |
- Exploring RAG pipelines and LLM fine-tuning for domain-specific applications
- Contributing to the open-source ML ecosystem
- Preparing research-oriented projects aligned with graduate programme applications
I am open to internships, full-time roles in Data Science / ML Engineering, and graduate programme collaborations.
📧 Reach me on LinkedIn or through my portfolio.
"Turning 100K+ rows of data into decisions that matter."