Skip to content
View ratulsaha2021's full-sized avatar
🎯
Focusing
🎯
Focusing
4 followers · 7 following

Block or report ratulsaha2021

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
ratulsaha2021/README.md

Hi there! I'm Ratul Saha | AI Researcher & Backend Engineer 👋

Typing SVG

🧠 AI & Deep Learning Focus

  • Neural Networks: Deep understanding of backpropagation, model production, and core mathematical foundations.
  • LLM & Transformers: Experienced in specialized Tokenizers, Encoder modifications, and lightweight architecture optimization.
  • Medical AI: Developed systems for Chest Disease Detection and Tuberculosis Diagnosis using CNNs and transfer learning.

⚙️ Backend & System Design

  • High-Performance APIs: Expert in FastAPI with a focus on "Fast Operation" and low-latency system logic.
  • Scalability: Proficient in Multi-manager authentication, Load Balancing, and Caching strategies.
  • Databases: Specialized in PostgreSQL Array storage and managing Data Redundancy through advanced schema design.

🛠 Technical Ecosystem

📊 Engineering Expertise Dashboard

Category Technical Level Key Specialization
Artificial Intelligence ██████████ 95% Core Neural logic & Transformers
Backend Architecture █████████░ 90% FastAPI, Scaling & Load Balancing
Database Engineering █████████░ 90% SQL Optimization & PG Arrays
System Security ████████░░ 85% Multi-manager Auth & Caching

🤝 Connect & Collaborate

Explore me: Ratul Saha | Portfolio

Refining the future through AI & robust engineering.

Pinned Loading

  1. bengali-llama-repo-chatbot bengali-llama-repo-chatbot Public

    This project is a RAG-style chatbot that answers questions about your LLaMA fine‑tuning repo: ratulsaha2021/Fine-Tuning-LLaMA-3.1-8B-for-Bengali-Empathetic-Dialogue.

    Jupyter Notebook

  2. ERP-Copilot-Prototype-with-Amazon_Product_Dataset ERP-Copilot-Prototype-with-Amazon_Product_Dataset Public

    This repository contains the design documentation and a functional prototype for integrating a Large Language Model (LLM) Copilot into an existing Enterprise Resource Planning (ERP) ecosystem.

    Jupyter Notebook

  3. Fine-Tuning-LLaMA-3.1-8B-for-Bengali-Empathetic-Dialogue Fine-Tuning-LLaMA-3.1-8B-for-Bengali-Empathetic-Dialogue Public

    Fine-tuning LLaMA 3.1-8B-Instruct to generate empathetic responses in Bengali using Parameter-Efficient Fine-Tuning (PEFT) and the Unsloth framework.

    Jupyter Notebook

  4. universal-RAG-API-kaggle universal-RAG-API-kaggle Public

    FastAPI-based RAG API that can ingest documents, build embeddings in FAISS, and answer questions via a /query endpoint.

    Jupyter Notebook

  5. Hospital-Management-System Hospital-Management-System Public

    Hospital Management System (HMS) is a database-driven solution designed to automate medical institution workflows. It features a robust relational schema for managing patients, staff, and billing. …

  6. Tuberculosis-Diagnosis-from-Chest-X-Ray-using-Deep-Learning Tuberculosis-Diagnosis-from-Chest-X-Ray-using-Deep-Learning Public

    Tuberculosis Diagnosis via Deep Learning is an IEEE-published AI solution that achieves 100% accuracy using EfficientNetB3 and DenseNet201. It features a web-based interface for real-time Chest X-r…

    Jupyter Notebook