About Me
Experienced Data Scientist with a master’s in computer engineering and 7 years in software development and data analytics. Specializes in predictive modeling, machine learning, deep learning, and AI-driven solutions, lev…
Experienced Data Scientist with a master’s in computer engineering and 7 years in software development and data analytics. Specializes in predictive modeling, machine learning, deep learning, and AI-driven solutions, leveraging OCR technologies, AWS, and Generative AI (LLMs) to transform complex data into actionable insights that drive business growth and efficiency.
Experience
Specialist- Software Engineering
<p>Designed and developed a comprehensive RESTful API platform for MongoDB and Redis database user life cycle management and access control automation.</p><p> Built a secure, scalable micro-services architecture using Flask-RESTful with JWT-based authentication, PostgreSQL request tracking and Swagger UI documentation.</p><p> Implemented automated workflows for user provisioning, role management and audit compliance across multiple database clusters.</p><p> Automated and developed secure RESTful API using Flask RESTful framework with JWT token-based authentication and role-based access control (RBAC).</p><p> Implemented comprehensive user lifecycle management including automated user creation, role assignment/revocation, account locking/unlocking and secure password generation with full audit trail capabilities.</p><p> Built asynchronous request system with PostgreSQL-backed request queue management, enabling both sync and async execution modes with real time status polling and error handling mechanism.</p><p> Integrated MongoDB access control with dynamic role extraction, privilege management and multi-VIP (virtual IP) support for centralizing database user administration across distributed MongoDB cluster.</p><p> Integrated Redis access control with dynamic acl rules extraction, privilege management and multi-VIP (virtual IP) support for centralize database user administration across distributed Redis multinodes.</p><p> Developed request tracking and audit logging system using PostgreSQL with SQLAlchemy ORM, capturing all user actions, request metadata, execution timeline and compliance data for regulatory requirements.</p><p> Designed modular architecture with separation of concerns-dedicated modules for views, functions, request management and Swagger documentation, ensuring maintainability and scalability.</p><p> Implemented error handling and validation with custom decorators for authentication, authorization JSON validation and standardized API response formats.</p><p> Created Swagger UI integration for interactive API documentation, enabling easy testing and onboarding API for consumers.</p><p> Established CORS policies and security measures including WAF ready configuration input validation with regex patterns and SQL injection prevention through parameterized queries.</p><p> Optimize database operations with connection pooling, efficient query patterns, and support for bulk operations across multiple VIP targets with partial success handling.</p>
Data Engineer
<ol><li>Designed and implemented a robust OCR pipeline to extract and structured financial data from bank statement pdfs, reducing manual efforts.</li></ol><p></p><ol><li>Designed a AI Clinical Knowledge Assistant to assist users with disease related queries by extracting information from uploaded PDF medical documents.</li></ol>
Associate Data Scientist
<p>Collected and pre-processed datasets using COCO, open images, and labelImg for annotation.</p><p> Implemented state-of-the-art object detection architectures such as YOLOv5 and faster R-CNN, leveraging pre-trained models for fine-tuning.</p><p> Trained the model using TensorFlow and OpenCV in a GPU powered Google Colab environment, optimizing hyperparameters for improved performance.</p><p> Evaluated performance using Precision, Recall, F1-score and Mean Average precision(mAP) for validation.</p><p> Designed and trained an XGBoost regression model for ETA prediction, incorporating historical travel times real-time traffic and weather conditions.</p><p> Implemented K-means clustering to categorize route into similar traffic patterns improving model performance.</p><p> Integrated Goggle maps API for real-time traffic updates and route optimization.</p><p> Reduced ETA prediction error by 25%, leading to better delivery scheduling and fuel savings.</p><p> Developed a FastAPI, deployed on AWS Lambda, handling real-time trip data.</p>
PROJECTS
1. Automated Data Extraction for Lending Platforms 2. AI Clinical Know
<p>Designed and implemented a robust OCR pipeline to extract and structured financial data from bank statement pdfs, reducing manual efforts.</p><p></p><p>Designed a AI Clinical Knowledge Assistant to assist users with disease related queries by extracting information from uploaded PDF medical documents.</p>