PI: Samuel AlbanieProject: Characterizing GPT-4 Capabilities for Remote Sensing through Systematic Profiling- Identified the GPT-4 knowledge capabilities for diverse factual and application based geographical tasks such as location query, information retrieval, route planning, etc.- Implemented prompt dialogs to fine-tune GPT-4 responses for general purpose knowledge queries.- Working towards creating a benchmark meta-dataset for evaluating LLMs on remote sensing tasks.

Boltu Robotics is an emerging autonomous delivery start-up. Here at Boltu, we are working on introducing reliable and large scale autonomous delivery with the help of state of the art Ai solutions in an exceedingly growing market. • Executed automated data labelling, filtering, and processing methods on AWS Sagemaker for large scale data collection.• Implemented and trained SOTA computer vision models using PyTorch for scene segmentation and object detection.• Worked on… Show more Boltu Robotics is an emerging autonomous delivery start-up. Here at Boltu, we are working on introducing reliable and large scale autonomous delivery with the help of state of the art Ai solutions in an exceedingly growing market. • Executed automated data labelling, filtering, and processing methods on AWS Sagemaker for large scale data collection.• Implemented and trained SOTA computer vision models using PyTorch for scene segmentation and object detection.• Worked on identifying and deploying vision models for automated lane detection and path prediction on an autonomous delivery robot Show less

- Worked on developing machine learning models for time series analysis and forecasting.- Processed realtime power station data, weather data, and energy price data to predict future energy demand and price using models like Linear Regression, XGBoost, LGBM, and Prophet. - Performed data cleaning, feature engineering, and time series visualization to prepare statistical insights using large scale energy demand data.- Developed regression models that could forecast future energy… Show more - Worked on developing machine learning models for time series analysis and forecasting.- Processed realtime power station data, weather data, and energy price data to predict future energy demand and price using models like Linear Regression, XGBoost, LGBM, and Prophet. - Performed data cleaning, feature engineering, and time series visualization to prepare statistical insights using large scale energy demand data.- Developed regression models that could forecast future energy demand and price with below 12% error rate. Show less