About

Archil Srivastava is a highly adaptable CS graduate student with a passion for software development and machine learning. With 3.28 years of experience in both industry and academia, Archil has a proven track record of success in various roles. Currently, he is an open-source contributor at Machine Learning for Science (ML4SCI), where he is working on the classification of dark matter substructure from strong gravitational lensing images. In this role, he optimized and benchmarked the performance of Hybrid Vision Transformers, achieving an AUROC of 99.6% using CoAtNet. The project will be used for the images captured by the Vera C. Rubin Observatory, which is currently under construction in Chile, and will help physicists better understand the structure of dark matter. Prior to this, Archil was a graduate student researcher at The Nemati Lab at UC San Diego, where he analyzed uncertainty estimation in deep learning-based sepsis prediction. He also engineered features for the prediction of culture results in septic patients using deep learning, which would circumvent the need for a lengthy bacteria culture. Archil presented this work at the American Medical Informatics Association (AMIA) Annual Symposium 2022. Archil's educational background includes a Master of Science in Computer Science from UC San Diego Jacobs School of Engineering, which he is expected to complete in 2023. He also holds a Bachelor of Technology in Computer Science from IIT, where he started his journey in the field. Archil is a research scientist, ML expert, software engineer, and deep learning specialist with a strong technical skill set. Archil is always eager to work with smart people on cool new technologies and solving challenging problems. He is a quick learner and has a keen eye for detail, making him an asset to any team. With his extensive experience and expertise in the field, Archil is poised to make a significant impact in the world of software development and machine learning.