Aryak Mohanty

Developed a real-time face recognition system using Convolutional Neural Networks (CNN) for secure authentication. The system comprises three modules: a data collection program that captures images only when humans are detected, a consolidation program to process the dataset, and a recognition model trained on this data to identify faces with high accuracy.

Implemented a high-accuracy spam email classifier using the Multinomial Naive Bayes algorithm. The model utilizes Natural Language Processing (NLP) techniques to analyze email content and effectively distinguish between legitimate messages and spam, enhancing email security.

Engineered an intelligent conversational AI chatbot with advanced Natural Language Understanding (NLU) capabilities for customer support applications. The system incorporates intent recognition, entity extraction, and context retention to enable coherent multi-turn dialogues. Implemented comprehensive safety guardrails and content moderation to ensure secure and appropriate interactions while maintaining professional conversation standards.

Developed a custom Google Search Engine web service using Node.js, Express, and the Google Custom Search API, featuring a secure environment-based configuration, a /search REST endpoint with query validation, API request logging, error-handling mechanisms, Axios-powered HTTP integration, and Ngrok tunneling support for external access; the system includes a technical debugging layer, a test automation script, and identification of issues such as missing frontend assets and strict dependency requirements, providing a robust and extensible backend architecture for search-driven applications.