Jayden Vasquez

Jayden Vasquez

Sophomore CSE @ Ohio State. I work on ML inference optimization, CUDA kernels, and edge deployment. Incoming AI intern at NASA Ames.

AI Agent Intern — NASA Ames Research Center Building AI agents for autonomous rover swarm coordination · Summer 2026 · Mountain View, CA
AI/ML Fellow — Cornell Tech (Break Through Tech) Developing and deploying ML projects, mentored by engineers from Google, Microsoft, and Meta
Career Prep Fellow — Management Leadership for Tomorrow Selective 18-month program · case studies, simulations, and executive mentorship
Sophomore @ The Ohio State University B.S. Computer Science and Engineering · GPA 3.68 · Morrill Distinction Full-Ride Scholar
AI Agent Intern — Rover Swarm Systems
NASA Ames Research Center
Jun 2026 – Aug 2026
  • Developing AI agent systems for autonomous rover swarm coordination on edge hardware
AI Systems Research Lead
AI Manufacturing Systems Lab · Ohio State
Jan 2025 – Apr 2026
  • Built real-time inference system for Honda partnership on constrained embedded hardware
  • Optimized DinoV2 ViT with a lightweight projection head for image classification
  • Deployed on NVIDIA Jetson AGX Orin with CUDA, cuDNN, and TensorRT
Autonomous Perception Engineer
Center for Automotive Research · Ohio State
Aug 2024 – Apr 2025
  • Architected real-time deep learning algorithms for autonomous vehicle perception using YOLOv11
  • Implemented distributed training across a multi-node SLURM cluster
  • Automated sensor feed wrangling with Pandas and NumPy
Custom CUDA Matmul Kernel

Three matrix multiplication kernels from scratch — naive, shared memory tiled, and double-buffered vectorized — integrated into PyTorch via C++ extensions and benchmarked against cuDNN. Profiled with Nsight Compute to isolate memory bandwidth vs compute bottlenecks at each step.

CUDA C++ PyTorch cuDNN Nsight Compute
Autonomous Drone Racing System

Vision-based system for the Anduril AI Grand Prix. Single FPV camera pipeline for gate detection without GPS. Real-time control loop over MAVLink/UDP at 50–120 Hz under strict compute constraints (~100 TOPS).

Python PyTorch CUDA C++ MAVLink
Languages
  • CUDA C++
  • Python
  • C / C++
  • Bash
ML / Inference
  • PyTorch
  • TensorRT
  • OpenVINO
  • cuDNN
Systems
  • Linux
  • Docker
  • AWS
  • SLURM
Tools
  • Nsight Compute
  • Git
  • OpenCV
  • NumPy / Pandas