Dhivya Sreedhar
Graduate student and ML researcher focused on deep learning
dhivya-sreedhar
Pittsburgh, Pennsylvania
Joined May 2026
Summary
Graduate student and active researcher in deep learning and machine learning, contributing to academic publications and course instruction at Carnegie Mellon University. google+2
Applies machine learning to practical environmental problems — notably building datasets and ML pipelines for plastic identification and sorting as part of a city pilot with Reclamation Factory. reclamationfactory+1
Combines academic research (including HCI and signal-based input research such as sEMG typing) with industry experience across ML and software engineering roles. arxiv+2
Experienced in both teaching and hands-on engineering: has served as a TA for CMU's intensive deep learning course and held multiple engineering/research positions spanning startups and established companies. cmu+1
Work
Education
Projects
Writing
Typing Reinvented: Towards Hands-Free Input via sEMG
November 1, 2025arXiv paper exploring surface electromyography (sEMG) as a non-invasive input modality for mapping muscle activity to keyboard inputs, using attention-based architectures and a lightweight decoding pipeline to enable real-time muscle-driven text input.
Turning the Plastic Problem into a Machine Learning Challenge
January 1, 2025A technical blog post describing dataset construction, augmentation strategies, model design (transfer learning and multi-head architectures), and evaluation for machine-learning-driven plastic identification and sorting applied in a city pilot.
Lane and Traffic Sign Detection in Self-Driving Cars using Deep Learning
January 1, 2024Research work on lane and traffic sign detection for autonomous driving using deep learning techniques (listed in Google Scholar author profile).