RESUME
RG
Rahul Gupta
Computer Vision Researcher & Full Stack Engineer
Computer VisionPyTorchTensorFlowCUDAFull StackDockerLLMsMLOps
TECHNICAL SKILLS
ML / AI
- •PyTorch & TensorFlow model development
- •Hugging Face transformers integration
- •Hyperparameter tuning & optimization
- •Model quantization techniques
COMPUTER VISION
- •OpenCV image processing pipelines
- •Object detection & segmentation
- •3D reconstruction algorithms
- •Neural Radiance Fields (NeRF)
GPU COMPUTING
- •NVIDIA CUDA infrastructure
- •Hybrid architecture (ARM-based H200, A100)
- •AMD-x86 optimization
- •WebGPU acceleration
MLOPS & DEPLOYMENT
- •Docker containerization
- •ONNX model conversion & optimization
- •CI/CD pipelines for ML workflows
- •Remote GPU management
EXPERIENCE
Computer Vision Researcher
2022 - 2024San Diego State University
- •Developed real-time object detection systems using PyTorch and TensorFlow
- •Implemented 3D reconstruction algorithms for spatial mapping applications
- •Optimized computer vision algorithms for edge devices, improving inference speed by 40%
Full Stack Engineer
2019 - 2022Walmart Labs, Sunnyvale
- •Architected and developed end-to-end web applications with React and Node.js
- •Implemented computer vision solutions for product recognition and inventory management
- •Containerized applications using Docker for consistent deployment across environments
- •Led development of GPU-accelerated image processing pipelines for content moderation
EDUCATION
M.S. Computer Science
2022 - 2024San Diego State University
Specialization in Computer Vision and Machine Learning
B.S. Computer Science
2014 - 2018San Diego State University
PROJECTS
Neural Radiance Fields
Browser-based NeRF implementation using WebGPU for 3D scene reconstruction.
Real-time Object Tracking
GPU-accelerated multi-object tracking system with occlusion handling.
WebLLM Integration
Browser-based large language model inference using WebGPU acceleration.
MLOps Pipeline for CV Models
Automated CI/CD workflow for training, testing, and deploying computer vision models.
