I build Machine Learning, NLP, and AI-powered applications and turn them into real-world web applications.
I'm particularly interested in Generative AI, Transformers, RAG, AI Agents, and intelligent developer tools.
- π€ Building AI & Machine Learning applications
- π§ Exploring Transformers, LLMs, RAG & AI Agents
- π Working primarily with Python for AI/ML
- π Building web applications with React & JavaScript
- β‘ Developing APIs with FastAPI and Node.js
- βοΈ Learning AI/ML deployment and cloud technologies
- π¬ Interested in NLP, Computer Vision & Generative AI
- π‘ Always experimenting with new AI technologies
An AI-powered code review system that automatically analyzes GitHub Pull Requests and provides intelligent code feedback.
- GitHub App integration
- GitHub Webhooks
- Pull Request analysis
- Automatic changed-file detection
- AI-powered code review
- Gemini integration
- FastAPI backend
- Automated workflow
Python FastAPI GitHub API Webhooks Gemini AI
π Repository: View Project
A document-based Question Answering system using BERT and Hugging Face Transformers.
Users can upload documents and ask questions based on the document content.
- π Document upload
- π§ BERT Question Answering
- π Context-based answers
- π€ Hugging Face Transformers
- β‘ FastAPI backend
- π Web interface
- βοΈ Cloud deployment
Python BERT Transformers FastAPI PyTorch Hugging Face
π Repository: View Project
A transformer-based text summarization application using T5.
The application converts long text into concise summaries using a fine-tuned/pre-trained Transformer model.
- π Text summarization
- π€ T5 Transformer
- β‘ FastAPI API
- π Web interface
- π Model deployment
Python T5 Transformers PyTorch FastAPI
π Repository: View Project
A Machine Learning application that predicts the likelihood of heart disease using Logistic Regression.
Accuracy: 90%
- Logistic Regression
- Data preprocessing
- Feature engineering
- Model evaluation
- Interactive Streamlit UI
- Real-time prediction
Python Pandas NumPy Scikit-learn Logistic Regression Streamlit
π Repository: View Project
A digital platform designed to securely organize important family documents and memories.
- π Document storage
- π Authentication
- π Digital document management
- β° Expiry reminders
- π‘ Property document organization
- π Certificate and insurance management
React JavaScript Supabase Tailwind CSS
π Repository: View Project
| Project | Algorithm / Technology | Result |
|---|---|---|
| β€οΈ Heart Disease Prediction | Logistic Regression | 90% Accuracy |
| π House Price Prediction | Regression | RΒ² ~0.65+ |
| π Customer Churn Prediction | Logistic Regression | ~79.5% Accuracy |
| π§οΈ Rainfall Prediction | Regression | Experimental |
| π’ MNIST CNN | CNN | 99%+ Accuracy |
| π’ MNIST RNN | RNN | ~96% Accuracy |
| π Text Summarization | T5 Transformer | NLP |
| π§ Document QA | BERT | NLP |
π€ Generative AI
β
π§ Large Language Models
β
π Retrieval Augmented Generation
β
πΈοΈ LangGraph
β
π€ AI Agents
β
β‘ Production AI Applications
Currently exploring:
- π€ Hugging Face Transformers
- π§ Large Language Models
- π RAG
- πΈοΈ LangGraph
- π€ AI Agents
- π Vector Databases
- β‘ FastAPI
- βοΈ AI/ML Deployment
- π³ Docker
π€ AI Applications
π§ Machine Learning Models
π Document Intelligence
π Question Answering Systems
π NLP Applications
π» AI Developer Tools
π Full-Stack Applications
π Production ML/AI Systems
Build β Experiment β Deploy β Learn β Improve π
I enjoy taking an idea from a machine learning experiment and turning it into a usable application.
I'm interested in collaborating on:
- π€ AI/ML projects
- π§ Generative AI
- π RAG applications
- π€ NLP projects
- π Full-stack AI applications
- π Open-source projects
If you find any of my projects interesting, feel free to β the repository and explore the code.
Let's build something intelligent together. ππ€