Srecharan Selvam
Graduate Research Assistant @Carnegie Mellon University Robotics Institute
Signup · Get unlimited contacts
WORK HISTORY
Graduate Research Assistant @Carnegie Mellon University Robotics Institute
Pittsburgh, PA, US
Developed a Vision Language Action (VLA) system through LoRA fine-tuning of LLaVA-1.5-7B (CLIP + Vicuna) foundation model, adapting multimodal reasoning from text generation to action prediction for robotic leaf manipulation.• Automated data creation via self-supervised learning pipeline, eliminating 100% manual annotation for GraspPointCNN.• Trained attention-based GraspPointCNN using MLflow to track 60+ model experiments for grasp point optimization.• Boosted inference (20→27 FPS) by parallelizing 3D projection with CUDA kernels & compiling models with TensorRT.• Deployed Dockerized VLA-enhanced grasping stack to a 6-DOF robot, achieving 82% leaf grasp success rate in field tests.
EDUCATION
DAV Group of Schools (TNAES), Chennai
High School Diploma
SSN College of Engineering
Bachelor of Engineering - BE
Carnegie Mellon University
Master of Science - MS
ABOUT SRECHARAN SELVAM
Hello, I\'m a graduate student pursuing my MS at Carnegie Mellon University with a specialization in machine learning, deep learning and computer vision. As a Graduate Research Assistant at CMU, I\'ve engineered a real-time vision system for autonomous plant manipulation that integrates stereo depth estimation with custom convolutional neural networks through a hybrid approach that combines classical computer vision with deep learning. My professional journey has taken me through diverse challenges - from designing real-time 3D hand tracking systems for AR interactions at Hanon Systems to implementing multi-camera vision systems for industrial automation at Vee Ess Engineering. Each role has strengthened my expertise in building robust ML systems that perform reliably in complex environments. I\'m fascinated by the capabilities of generative models and Generative AI, having benchmarked GANs, VAEs, and diffusion models for image synthesis, and built end-to-end systems leveraging NLP and time-series analysis for quantitative stock trading. What drives me is creating ML systems that bridge theoretical performance with practical application. I excel at data acquisition and preprocessing, and I\'m proficient with PyTorch, TensorFlow, OpenCV, and scikit-learn, leveraging these frameworks to develop scalable solutions that deliver real value across industries. My work combines technical rigor with a focus on impact, consistently pushing the boundaries of what\'s possible in applied machine learning.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.