CV

Education

• BSc in Computer Science and Engineering, Rajshahi University of Engineering and Technology, 2023
  • Coursework: Linear Algebra, Data Structures and Algorithms, Object Oriented Programming, Discrete Mathematics, Database Management, Applied Statistics & Queuing Theory, Digital Image Processing, Neural Network and Fuzzy System, Artificial Intelligence, Data Mining

Work experience

• Data Scientist: Manaknightdigital
  • Machine Learning
  • Deep Learning
  • Data Analysis & Visualization
  • Experience with Statistical Tools
  • Written & Verbal Communication
  • Scripting Language: Python
  • Database: MySQL,MongoDB,Sqlite3
  • ML Algorithms and Deep Learning Algorithms
  • NLP: Text Normalization
  • BERT,Sentiment Analysis,Word Embedding
  • Data Visualization & Analysis: PowerBI, Excel
  • Web Framework: Flask , Libraries: Tensorflow , Scikit-learn , Keras , Seaborn
  • Time Series Analysis
  • Large Language Models(LLMs)

Projects:

• Chatbot Development:
  • Collected and processed product information using Excel, pandas, and openpyxl.
  • Integrated GPT-4 to respond to user queries and manage token size limitations.
  • Utilized libraries like nltk, sklearn, and Flask for deploying the chatbot.
• Fraud Detection System:
  • Performed EDA and feature extraction on transaction datasets.
  • Developed and optimized ML models including Xgboost, SVC, and Logistic Regression.
  • Achieved 90% accuracy in detecting fraudulent transactions and deployed the system using Flask.
• Data-driven ChatBot for Financial Queries:
  • Implemented RAG and Pinecone, enhancing data retrieval speed by 40%, enabling faster decision-making for lenders.
  • Improved data retrieval accuracy by 25% using Corrective RAG Approach.
  • Applied Beautiful Soup and PyPDF2 for data scraping and processing.
• Sports Data Analysis ChatBot:
  • Scraped and analyzed football data to predict match outcomes.
  • Integrated RAG and Pinecone for efficient data querying and vector database management.
  • Employed Beautiful Soup and PyPDF2 for data collection, analyzing 2 million football data points to achieve a 90% prediction accuracy, supporting strategic betting decisions.
• Custom Image Generation System:
  • Developed an image generation platform using Stable Diffusion.
  • Trained custom models to generate images based on user-defined presets.
  • Utilized PyTorch and transformers for model training and deployment and finally used Docker for containerization.
• AI-driven Data Matching System:
  • Chunked organizational data into sections using models like Llama-2-7B.
  • Applied cosine similarity for matching data to specific tenders.
  • Integrated GPT-4 for generating insights from matched data.
  • Matched organizational data against specific tenders, increasing successful tender submissions by 70%.

Skills

• Machine Learning
• Deep Learning
  • Computer Vision
  • Natural Language Processing(NLP)
  • Object Detection
• Data Mining
• Statistics
• Artifitial Intelligence
• Data Structure & Algorithm
• Database Management System
• Neural Network and Fuzzy System
• Generative AI

Personal Projects:

• Decoding AI vs Human:
  • Goal: Decoding AI vs Human is an interactive web application that allows users to put any text and see if a human or an AI wrote it. This application is trained on the Machine Hack dataset.
  • Library/Technology: scikit-learn, AWS
• PineconePDFExtractor PyPi:
  • Goal: PineconePDFExtractor is a Python library for extracting text from PDF files for pinecone.
  • Library: PyPdf2
• DataSciencePilot GitHub
  • Goal: It is a chat-based interface designed to interact with custom PDF files. It leverages the power of Pinecone for efficient vector database management and LLaMA-2 for advanced query response capabilities.
  • Library: Pinecone, Langchain, Transformers
• CVAnalyzerPro StreamlitApp
  • Goal: matches participant’s CVs with the company’s requirements and gives scoring.
  • Library: openai, Gemini, Streamlit
• CaptionCraft StreamlitApp
  • Goal: generate caption using Google Gemini API.
  • Library: Gemini, Streamlit
• PredictStock StreamlitApp
  • Goal: to predict the stock of any company like Google, Microsoft, Apple.
  • Solution: used LSTM to train the model.
  • Library: Tensorflow, Streamlit
• Diabetes Prediction GitHub
  • Goal: predict whether any patient has diabetes or not.
  • Solution: used Artificial Neural Network(ANN) to train the model and predict the disease.
  • Library: PyTorch, Flask, Gunicorn
• Movie Recommendation
  • Goal: recommend the movie based on the movie entered by user.
  • Solution: Used KNN to find the nearest 5 movies using cosine similarity.
  • Library: pandas, numpy, sklearn, Flask, scipy
• Market Price Prediction GitHub
  • Goal: to predict the price of the product using ARIMA, SARIMAX, LSTM, FbProphet, GRU, Xgboost.
• Potato Disease Classification Using CNN GitHub
  • Goal: To classify disease in Potato.
  • Solution: the dataset is taken from Kaggle which contains 1506 images with 3 classes.
  • Result: Overall accuracy is 100%.
  • Library: Tensorflow

Publications

An Attention-Based Deep Learning Approach to Knee Injury Classification from MRI Images

K. D. Nath, A. F. M. M. Rahman and M. A. Hossain "An Attention-Based Deep Learning Approach to Knee Injury Classification from MRI Images." IEEE. 1(1).

Download my resume

Kowshik Deb Nath Resume