I am a Data Scientist passionate about Transforming data into actionable insights and solving real-world problems. With expertise in Machine learning, Deep Learning, NLP, Predictive analytics, Generative AI, and feature technology, I build innovative solutions to drive impactful outcomes. My mission is to harness data to empower decisions and create meaningful change across industries.
- Programming Languages: Python, SQL
- Databases: MySQL, PostgreSQL, MongoDB
- Big Data Tools: Hadoop, Spark
- Data Analysis: Pandas, NumPy, Statsmodels, Matplotlib, Seaborn, Plotly
- Machine Learning: Scikit-learn, XGBoost, LightGBM, CatBoost, MLflow
- Deep Learning: TensorFlow, Keras, PyTorch
- Natural Language Processing (NLP): NLTK, SpaCy
- Computer Vision (CV): OpenCV
- Generative AI: GPT-4, Llama, BERT, RAG, Transformers
- Cloud Platforms: AWS, Google Cloud Platform
- Version Control: Git, GitHub
- Frameworks: Streamlit
- IDEs: Google Colab, Jupyter Notebooks, VS Code
-
AI Healthcare Assistant: Developed an AI-powered tool for healthcare professionals, integrating a chatbot, Retrieval-Augmented Generation (RAG), and medical diagnosers with high accuracy across domains like cardiology and pulmonology. Improved diagnostic efficiency and decision-making for over 10,000 cases.
-
AI Finance Assistant: Built an intelligent tool for financial sentiment analysis using FinBERT for domain-specific accuracy and Llama for nuanced insights. Provides real-time sentiment analysis of financial news and reports, empowering investors and analysts to make informed decisions.
-
Retrieval-Augmented Generation Application: Built an AI-driven tool using Meta Llama 3.1 to enhance decision-making by retrieving relevant data and generating accurate, actionable insights. It streamlines workflows, improves accuracy, and aids professionals in making better, faster decisions.
-
Customer Conversion Prediction Application: Developed a Random Forest Classifier model with 91% accuracy to predict customer insurance subscriptions, leveraging data preprocessing, EDA, and real-time deployment for improved marketing targeting.
-
YouTube Data Harvesting and Warehousing Application: Built a Streamlit application to collect and analyze YouTube channel data using Google API, achieving seamless integration with MongoDB for data storage and SQL for structured querying, enabling users to retrieve and visualize metrics from multiple channels efficiently.
-
PhonePe Pulse Data Visualization Application: Developed an interactive geo-visualization dashboard using Streamlit and Plotly to analyze PhonePe Pulse data, integrating Python and MySQL for efficient storage, retrieval, and dynamic user insights with over 10+ customizable dropdown options.
-
BizCardX: Extracting Business Card Data With OCR Application: Developed a Streamlit-based application leveraging easyOCR to extract business card details, including company name, contact information, and address. Integrated SQL for database management, allowing users to read, update, and delete entries via an intuitive GUI, enhancing business card organization and accessibility.
- Jr. Design Requirement Analyst at Vectra Automation (Nov 2023 - Present)
- Worked on analyzing design requirements and coordinating with cross-functional teams to ensure alignment with project goals.
- Master of Data Science from IIT GUVI
- Bachelor of Engineering in Mechanical Engineering from Nadar Saraswathi College of Engineering and Technology
