Divya . Email & Phone Number

Trenton, Nj, Us

I developed the Fraud Detection System for Secure Transactions to prevent fraud in real-time. The backend, built with Python and Django, ensured security and scalability, while PostgreSQL managed structured data such as user profiles and transactions. AI/ML models using scikit-learn and TensorFlow were integrated to analyze transaction patterns and detect anomalies, enabling immediate fraud detection. These models were trained on historical data and optimized through AWS SageMaker, using AutoML tools to enhance accuracy.The frontend, developed with React.js, HTML, CSS, and JavaScript, provided users with a seamless interface for real-time transaction tracking. AJAX enabled efficient, real-time data updates, allowing users and security teams to monitor and address fraud instantly. The system used PostgreSQL for structured data, MongoDB for unstructured logs, and AWS DynamoDB for real-time access to fraud history. AWS Lambda supported serverless scaling, while AWS CloudFormation automated infrastructure deployments.DevOps practices utilized Jenkins, Docker, and Kubernetes for CI/CD, ensuring the system was consistently updated and capable of deploying new AI/ML algorithms without downtime. Snowflake was employed for data warehousing, with Apache Spark handling large-scale data processing. Combined with CloudWatch and SNS for monitoring and alerts, this setup created a scalable, AI-powered fraud detection system.

I developed a Personalized Recommendation System to enhance user experience and boost sales through AI-powered suggestions. This system utilized advanced AI/ML models to deliver tailored product recommendations based on user behavior, purchase history, and browsing patterns, significantly improving Etsy’s customer engagement.The backend was built with Python and Django, incorporating a microservices architecture for scalability. TensorFlow, scikit-learn, and Keras powered the recommendation models, employing collaborative filtering, content-based filtering, and deep learning algorithms for accurate predictions. AWS SageMaker was used for model training and hyperparameter tuning, while AutoML optimized performance. Data storage was managed through MySQL for structured data and AWS DynamoDB for real-time access to user interaction logs.For real-time interactions, I created a responsive React.js frontend that provided instant product recommendations. AJAX facilitated seamless communication between the frontend and backend, ensuring smooth user experiences. The project employed AWS Lambda, S3, and API Gateway for a serverless architecture, offering scalable solutions during peak traffic. Redis was utilized for data caching, enhancing performance for frequent recommendation queries.Infrastructure was managed with Docker and Kubernetes for containerization and orchestration, ensuring efficient deployment. Jenkins enabled CI/CD pipelines for continuous updates of AI/ML models, while Terraform facilitated infrastructure as code (IaC) for consistent deployments. For monitoring and alerting, I integrated Amazon CloudWatch, Prometheus, and Grafana to track system health, with Amazon SNS for notifications on critical updates. This comprehensive approach resulted in a highly scalable, automated, data-driven recommendation system that successfully increased user engagement and sales conversions.

In my role, I spearheaded the design and development of a sophisticated client-side application, leveraging an extensive array of cutting-edge technologies. Utilizing Python, JavaScript, React, HTML, CSS, and incorporating Pandas and SciPy for efficient data processing, I crafted the backbone of the system. Flask served as the robust web framework for building scalable applications, while I ensured seamless communication by implementing a microservices architecture with Flask for backend services and Node.js for frontend services.To address scalability concerns, I integrated AWS services such as S3 and EC2 during deployment. I seamlessly incorporated cloud databases, including RDS and DynamoDB, in alignment with DevOps practices for continuous integration and continuous deployment (CI/CD). Kubernetes managed container orchestration, while Jenkins and Docker facilitated continuous integration and deployment processes.For version management and collaborative coding, I proficiently utilized Git and GitHub. Thorough testing procedures were meticulously executed using pytest to guarantee the reliability and performance of the system.Key Skills and Technologies:Programming Languages: Python, JavaScriptWeb Development: React, HTML, CSS, FlaskData Processing: Pandas, SciPyMicroservices Architecture: Flask, Node.jsAPI Development: Express.jsCloud Services: AWS (S3, EC2, RDS, DynamoDB)DevOps: CI/CD, Docker, KubernetesVersion Control: Git, GitHubTesting Tools: pytest