Developing AI-driven solutions to improve wildlife monitoring across the United States. By integrating AI into our workflow, we're reducing the need for manual processing and improving classification across large datasets. One of the key aspects of our work involves depth estimation models for calibrated data, processing over 100,000+ camera trap images from diverse ecosystems. Our system improves detection accuracy by 20-30% compared to traditional methods, allowing researchers to extract precise positional data from images captured across 50+ monitoring sites nationwide. Automating these processes has resulted in a 40% reduction in manual labor, ensuring more consistent and reliable wildlife monitoring while accelerating data-driven conservation efforts.

Developed machine learning models to predict thermal conductivity profiles with improved accuracy. Using MATLAB's API Engine with Python, I generated over 30,000 data points, reducing manual processing time by 50% and enhancing model efficiency. To optimize predictions, I implemented Regression algorithms and ensemble learning, improving accuracy by 25% and reducing computational time by 30%. These advancements made thermal modeling more scalable and precise, enabling faster and more reliable simulations. Beyond model development, I refined data processing workflows, accelerating simulation speed by 40% and ensuring a more automated and efficient approach to material behavior analysis.

Worked on improving the platform frontend and integrating API-driven features. Implemented responsive design using HTML5, CSS3, and JavaScript frameworks and contributed to RESTful API development for real-time profile booking management.