Graduate Research Assistant

University of Illinois Urbana-Champaign

September 2024 - Present

• •Designed an AI-driven tool for analyzing human behavior in videos, utilizing GPT-4's multimodal capabilities to derive clinically relevant insights
• •Developed an interactive ReactJS-based UI with AI-generated visualizations and textual summaries to enable effective interpretation of results
• •Engineered a backend with Flask for real-time video data processing and GPT-4 inference, optimized for low latency responses

Software Engineer Intern

GlobalLogic Inc.

June 2024 - July 2024

• •Developed a real-time gamified quiz module, serving over 1000 users simultaneously and boosting user engagement by 25%
• •Architected a scalable microservices architecture, leveraging React.js, for frontend, Node.js for APIs and Firebase Functions for serverless compute
• •Optimized performance with a distributed Redis caching layer, reducing database load by 40% and improving response times by 35%
• •Integrated secure authentication with OAuth 2.0 (Google SSO), and implemented role-based access control using JSON Web Tokens (JWT)
• •Automated CI/CD pipelines with GitHub Actions and Docker, accelerating feature deployment and minimizing downtime
• •Built an LLM-based question-answering chatbot for the learning platform using GPT-4 API, delivering accurate, context aware, low-latency responses
• •Engineered a Retrieval Augmented Generation (RAG) pipeline using LangChain, leveraging Pinecone vector database for semantic embedding search

Software Engineer

Dell Technologies

July 2024 - December 2024

• •Developed a computer vision model using Python and PyTorch to identify product label defects in real-time
• •Utilized state-of-the-art CNN models for text localization and blur detection, and performed hyperparameter tuning to ensure robust performance
• •Built an event-driven data pipeline using Kafka, processing over 10,000 images daily for automated data ingestion and processing
• •Deployed the solution with Docker and Kubernetes and built a RESTful API using Python FastAPI for integration with the manufacturing workflow
• •Collaborated cross-functionally with engineers, designers and production managers to align model performance with production standards
• •Participated in code reviews, planning, and estimation meetings, ensuring timely delivery of features and adherence to best practices

Software Engineer Intern

Dell Technologies

January 2024 - June 2024

• •Designed an inventory management dashboard to monitor and visualize key performance indicators (KPIs) for the manufacturing supply chain
• •Built a responsive Angular frontend with advanced search and filtering, integrated with Java Spring Boot APIs and PostgreSQL database
• •Performed comprehensive data analysis of inventory, production and sales data using Pandas to identify features that impact product shortages
• •Developed a time series forecasting model using an LSTM network in TensorFlow, achieving improved demand and shortage predication
• •Built an automated, scalable ETL pipeline using Apache Airflow, reducing data processing time and ensuring fast model inference
• •Enabled interactive data visualization using charting libraries and optimized database queries for fast retrieval to ensure a smooth user experience
• •Implemented a real-time notification system using Web Sockets and RabbitMQ, reducing user response time to critical events by 50%

Machine Learning Engineer Intern

Cognida.ai

September 2022 - December 2022

• •Developed a computer vision system using OpenCV and PyTorch to track individual customer movements in supermarkets in surveillance video feed
• •Implemented the DeepSORT algorithm for multi-object tracking across camera views, and applied image inpainting to deal with occlusions
• •Created visualizations of the collected data to provide insights into customer behavior and enable optimization of product placement

Achievements