Skip to content
View Prithwijit24's full-sized avatar
💭
I may be slow to respond.
💭
I may be slow to respond.

Block or report Prithwijit24

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Prithwijit24/README.md

Hi, I'm Prithwijit Ghosh 👋

🧑‍💻 Data Scientist @ Accenture Technology  ·  🎓 M.Sc. Statistics, IIT Kanpur

📈 Forecasting  ·  ⚠️ Risk Scoring  ·  🚨 Anomaly Detection  ·  ⚙️ MLOps  ·  🤖 Applied ML & AI Systems


Portfolio Resume LinkedIn Email


About

I build applied ML and AI systems end-to-end — from $\color{orange}{\text{statistical modeling}}$ and forecasting to $\color{skyblue}{\text{agentic AI pipelines}}$ and production deployment. Currently working as a Data Scientist at Accenture Technology, with a background in Statistics from IIT Kanpur. I care less about the buzzwords and more about whether the thing I build actually ships, scales, and holds up under real-world data.

  • 🔭 Currently building agentic RAG systems, $\color{orange}{\text{forecasting models}}$, and risk-scoring pipelines
  • 🧠 Grounded in statistics — comfortable moving between classical models and modern $\color{skyblue}{\text{deep learning}}$ / LLM-based systems
  • ⚙️ Interested in taking models from notebook to production: APIs, containers, and CI/CD — not just accuracy metrics
  • 📫 Reach me at ghoshprithwijit39@gmail.com

Tech Stack

Languages & Data

Languages & Data

Machine Learning

Machine Learning

GenAI & AI Systems

GenAI & AI Systems

Engineering & MLOps

Engineering & MLOps

Analytics & Tools

Analytics & Tools

GitHub Activity

Numbers only tell part of the story, but they're a useful starting point. Below is a live snapshot of how I actually work day to day — total $\color{orange}{\text{contributions}}$ across public and private repositories, the languages I reach for most often when building models and pipelines, and how consistently I show up to commit code. I care more about shipping working systems than chasing green squares, but tracking this activity keeps me honest about momentum over time. It's regenerated automatically every night, so what you're seeing here is never stale — always a $\color{skyblue}{\text{current reflection}}$ of ongoing work.

These cards refresh automatically every day through a scheduled $\color{orange}{\text{GitHub Actions}}$ workflow, pulling fresh data straight from the GitHub API — no manual updates, no stale screenshots, just an always-current view of recent activity across all my repositories.




Statistics at the foundation. Machine learning in practice. AI at the edge.

Pinned Loading

  1. agentic_travel_planner agentic_travel_planner Public

    An agentic travel planner that combines LangGraph, hybrid Graph + Vector RAG, Neo4j, and ChromaDB to build context-aware travel itineraries. It uses multiple agents for planning and critique, with …

    Python

  2. skinwise_intelligent_recommender skinwise_intelligent_recommender Public

    An AI-powered skincare recommendation system combining facial attribute prediction with an agentic product research pipeline. It uses FaceNet embeddings and classical ML models for age, gender, and…

    Python

  3. density_power_divergence density_power_divergence Public

    A robust regression study using the Density Power Divergence estimator for multiple linear regression, with simulation experiments, efficiency analysis, outlier detection, and real-data application…

    R

  4. music_recommendation music_recommendation Public

    A production-oriented music recommendation system built around content-based retrieval from the Million Song Dataset, compressed audio embeddings, and nearest-neighbor search. The architecture also…

    Python

  5. credit_card_fraud_detection credit_card_fraud_detection Public

    A production-grade credit card fraud detection platform with leakage-safe feature engineering, DuckDB feature storage, PR-AUC optimized model training, Kafka-based micro-batch scoring, FastAPI serv…

  6. prithwijit_ghosh_portfolio21 prithwijit_ghosh_portfolio21 Public

    Personal portfolio built with React and TypeScript, featuring interactive visualizations and an RAG-powered AI chatbot.

    TypeScript