I'm a Data Scientist based in Colchester, UK, with an MSc in Data Science (Distinction) from the University of Essex. I build end-to-end machine learning and deep learning solutions — from data pipelines and modeling to deployment — across healthcare, signal processing, and applied AI.
- 🎓 MSc Data Science (Distinction), University of Essex — Jan 2026
- 🎓 BSc Computer Science & Engineering (Cum Laude), North South University — Jan 2024
- 📍 Colchester, Essex, UK — Right to work in the UK (Graduate visa, valid until Jul 2028), no sponsorship required
- 🔭 Currently exploring Data Scientist / ML Engineer roles
Languages & Core: Python, SQL (PostgreSQL, SQLite), Pandas Analytics & BI: Power BI, DAX, Excel, star-schema modelling, A/B testing ML/DL: TensorFlow, PyTorch, Scikit-learn, XGBoost, LightGBM, CatBoost, LSTM, Transformers NLP & Signal Processing: NLP, EEG/signal processing (MNE-Python) Deployment & Tools: Flask, Docker, REST APIs, MLflow, AWS, GCP Mobile: Flutter
| Project | Description |
|---|---|
| 🏠 UK House Price Prediction | Predicts sale prices from 2.7M real Land Registry sales; time-based validation, leakage-safe target encoding, gradient boosting vs baselines |
| 🏗️ UK Property Price Pipeline | End-to-end pipeline: Python cleaning and validation, SQL star schema, Power BI dashboard spec, CI-tested |
| 🏥 NHS A&E Performance Analytics | SQL star schema, Excel dashboard and Power BI model on 16 years of real NHS England data |
| 🗺️ UK Regional Inequality (Power BI) | Power BI star-schema project on ONS, Land Registry and MHCLG public data |
| 🩸 CGMacros: Multi-Horizon Glucose Prediction | Multimodal ML (CGM signals, nutrition, biomarkers, food-image embeddings) forecasting post-meal glucose spikes with LSTM, Transformer & ensemble models |
| 🧠 EEG Mental Workload Classification | Classifies mental workload from EEG signals — CNN reaches 93.6% accuracy vs. Logistic Regression & Transformer baselines |
| 🌾 AgroVision — AI for Precision Farming | Flutter app bringing on-device AI (TensorFlow Lite) to farmers: soil classification, crop recommendation, weed & yield detection |
| 📊 Customer Churn Prediction | End-to-end churn modeling with feature importance and model comparison |
| 🧪 A/B Testing Analysis | Statistical analysis of an A/B test — conversion rates, confidence intervals, funnel & segment breakdowns |
| 📈 Sales Dashboard | Interactive Streamlit sales analytics dashboard |
See all repositories for more.
📧 pulak2164@gmail.com · 💼 LinkedIn