Headline

Data Analyst | ML • Python • Power BI | Turning Data into Decisions | Open to Work

About

I don't just build models — I build the systems around them. From a multi-source ETL pipeline powering a 91%-accurate energy demand forecaster, to a customer segmentation engine that classified 1,000+ OTT users with 94% accuracy — every project I've shipped starts with messy real-world data and ends with something a non-technical stakeholder can actually use. What I do: ▸ Design end-to-end ML pipelines (data ingestion → feature engineering → model training → deployment) ▸ Tune ensemble models (XGBoost, LightGBM, CatBoost, Random Forest) to production-grade accuracy ▸ Build full-stack interfaces that make model outputs accessible — not just notebooks ▸ Translate complex analytical results into decisions that drive business action My stack: Python · SQL · Scikit-learn · PyTorch · XGBoost · LightGBM · Power BI · Tableau · MySQL · PostgreSQL · AWS · Git What makes me different: I'm a rare hybrid — I understand the math behind the models AND can ship the front-end interface that puts predictions in front of users. My portfolio includes a custom-built AI chatbot (LLM-integrated, deployed via Netlify with CI/CD) — because AI should be usable, not just impressive in a notebook. Certified in Generative AI and Prompt Engineering (IBM · Vanderbilt University via Coursera). AWS Cloud Fundamentals certified. BTech IT @ KIIT Bhubaneswar — CGPA: 8.11, 2026. Currently looking for: → Machine Learning Engineer · Applied AI Engineer · Data Analyst · Forward Deployed Engineer 📩 ssarkar4483@gmail.com

Location

Kolkata

AI stack

gemini

github copilot

hugging face

openai

Education

Bachelor of Technology, Information Technology, KIIT - Kalinga Institute of Industrial Technology

High School Diploma, CBSE, Kalyani public school