Trisha Sengupta

Implemented message queuing architecture using RabbitMQ for communication between microservice ,with retry mechanisms, message acknowledgment, and dead-letter queues to ensure reliable and fault-tolerant processing Developed API endpoints for authentication systems, including secure user registration, login, and session management, supported by well-structured database schema design Implemented DevOps practices by containerizing services using Docker, setting up CI/CD pipelines for automated build and deployment, and monitoring application performance to ensure reliability and scalability

Backend Development Intern at Advertyzement (2025-10 – Present)

• Implemented message queuing architecture using RabbitMQ for communication between microservice,with retry mechanisms, message acknowledgment, and dead-letter queues to ensure reliable and fault-tolerant processing
• Developed API endpoints for authentication systems, including secure user registration, login, and session management, supported by well-structured database schema design
• Implemented DevOps practices by containerizing services using Docker, setting up CI/CD pipelines for automated build and deployment, and monitoring application performance to ensure reliability and scalability

Machine Learning Intern at NIT Durgapur (2024-06 – 2024-08)

• Engineered and deployed a 10-genre music classification system utilizing Convolutional Neural Networks (CNN) and Transformer architecture, achieving 93% accuracy
• Performed comprehensive Exploratory Data Analysis (EDA) on audio datasets and implemented a feature extraction pipeline comparing Mel-Spectrograms and Mel-Frequency Cepstral Coefficients (MFCC), resulting in an 8% accuracy improvement through optimal feature selection
• Implemented a data-aware LLM assistant using RAG, enabling real-time question answering over custom datasets through document retrieval and contextual generation

– Heritage Institute of Technology

– Delhi Public School