Mathews P Mathew

Mathews P Mathew

Data Scientist

📍 Kochi, Kerala

a human being.


Tech Stack

PythonSQLPandasNumPy MatplotlibPower BITableauPlotly Machine LearningXGBoostLangChainLangGraph RAG SystemsFAISSGemini API AWSGCPDockerKubernetes AirflowFastAPIStreamlitBigQuery FigmaAdobe Suite

Projects

GuideForCoach — AI Football Scouting System

Built a KNN-based scouting tool; created a LangGraph multi-agent system (ML + Tavily + Gemini) deployed via FastAPI.

LangGraphFastAPIKNNGemini

Diabetes Chatbot Assistant — RAG Healthcare

RAG system using LangChain and Gemini API for fact-checked medical responses. Semantic search with Sentence Transformers & FAISS, deployed on Streamlit.

LangChainRAGFAISSStreamlit

Automated Data Pipeline & Reporting — AWS

Automated AWS pipeline (Glue, Lambda) for streamlined querying. Built secure QuickSight dashboards for real-time NoSQL insights.

AWSLambdaGlueQuickSight

Customer Churn Prediction — ML

End-to-end ML pipeline — trained multiple models and achieved 79% accuracy with XGBoost after hyperparameter tuning via GridSearchCV. Deployed on Streamlit.

XGBoostScikit-learnStreamlit

Questioning PDF — RAG on Streamlit

Uploaded PDFs split into 800-character chunks with overlap for context continuity. Saved into ChromaDB and prompted via gemma-3-1b-it.

ChromaDBStreamlitRAG

Automated Daily Sales Pipeline — GCP

GCP pipeline: data uploaded to bucket → BigQuery via Airflow → analyzed with a scheduled SQL view → visualized in Looker Studio.

GCPBigQueryAirflowLooker Studio

Journey

Aug 2024 → Present

Brototype Kochi

Data Science intensive program.

2022 — 2023

GDSC Lead — Google Developer Student Clubs

Led a team of 12, organized workshops on Google technologies.

2022 — 2023

IEEE Design Team — SJCET & Kochi Subsection

Lead Designer for IEDC Summit '22 and core team for GDSC WOW 2023.

2020 — 2024

BTech in Computer Science

St. Joseph's College of Engineering and Technology, Palai.