Remote Frontend Developer - Boston, Massachusetts

Cogito - AI Brainstorming Assistant using OpenAI API, Next.js 14, Tailwind CSS, Firebase, and Google Cloud Platform

- Developed "Cogito," an AI brainstorming assistant utilizing the OpenAI API to assist users in generating creative ideas and solutions.

- Created a dynamic and responsive interface with Next.js 14 and Tailwind CSS, ensuring an intuitive user experience.

- Implemented real-time data storage and user authentication using Firebase, and leveraged Google Cloud Platform for scalable back-end services, enhancing performance and reliability.

Contactless Lie Detection with Machine Learning, Kafka, and SQL

• Developed a contactless lie detection system utilizing machine learning and deep learning models, achieving a

70% accuracy rate in identifying deception based on facial and behavioral cues.

• Established an ETL pipeline with Python, Kafka, and SQL to ensure the accuracy and completeness of data,

enhancing the reliability of the system.

• Employed computer vision and machine-learning classification algorithms to analyze facial expressions and other

nonverbal cues from 20 participants, enhancing the system's robustness.

University Management System Chatbot using Python and Neural Networks

• Designed and implemented a Python Flask-based chatbot for the university management system, incorporating

deep learning techniques and PyTorch.

• Implemented a sophisticated system enabling the chatbot to automatically comprehend and respond to user

inquiries, leveraging artificial neural networks with an error rate of 0.02.

• Collaborated with a team of 3 and engaged university staff to gather comprehensive user requirements.

Stock Prediction using Historical Data with Machine Learning and Azure Cloud

• Developed a Random Forest machine-learning model for stock prediction analysis using Python, effectively

deploying it to the Microsoft Azure Cloud platform for scalability and performance.

• Conducted comprehensive data analysis and performed feature engineering to optimize the model's accuracy,

resulting in a notable 20% reduction in prediction errors.

• Leveraged the power of Azure Cloud to streamline data processing and model training.

Customer Segmentation using Python and Machine Learning

• Conducted in-depth analysis of mall customer data consisting of 200 customers using Apache Pig, extracting

features for customer segmentation.

• Implemented the K-means clustering algorithm in Python to group mall customers based on their purchasing

behavior and demographics, facilitating targeted marketing strategies.

• Utilized the identified customer segments to develop personalized marketing initiatives, such as tailored product

discounts and targeted advertising campaigns.