FFFF
Skip to content
View justin-gahona's full-sized avatar

Block or report justin-gahona

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
justin-gahona/README.md

Hey, I'm Justin Gahona

B.S. Computer Engineering → M.S. Electrical Engineering @ Texas Tech University
Texas Instruments SDE Program
19 years old | Minneapolis, MN → Lubbock, TX
Open to FPGA, embedded systems, firmware, and embedded ML roles

Projects

ECG Arrhythmia Classifier

Real-time 1-D CNN deployed on two independent embedded targets — STM32F446RE via X-CUBE-AI and a Nexys A7-100T FPGA via a hand-written SystemVerilog inference pipeline. Trained on MIT-BIH Arrhythmia Database. 98.17% INT8 accuracy (float32 baseline 98.14%). Added adaptive average pooling to reduce FC1 parameters 11× and fit within the STM32F446RE's 512 KB flash constraint. FPGA target renders live scrolling ECG + classification label on VGA at 640×480.

PyTorch SystemVerilog STM32 FPGA C Embedded ML

→ View Repository


Custom 16-bit RISC CPU with Hardware GPU

Designed from scratch in SystemVerilog on Basys 3 (Artix-7) FPGA. Custom ISA, ALU, register file, memory-mapped GPU with 32-sprite renderer, SPI NOR Flash cartridge interface, and Python assembler. Synthesized to 3,448 LUTs at 100 MHz. Runs Pong end-to-end as the demonstration application.

SystemVerilog FPGA CPU Architecture Vivado

→ View Repository


Skills

HDL: SystemVerilog, Verilog
Embedded: STM32, C, UART, SPI, DMA
ML: PyTorch, ONNX, INT8 Quantization
Tools: Vivado, STM32CubeIDE, X-CUBE-AI, KiCad, Git

🔗 LinkedIn

Popular repositories Loading

  1. justin-gahona justin-gahona Public

    Profile README for Justin Gahona – ECE @ Texas Tech, SDE Program

  2. ecg-arrhythmia-classifier-stm32 ecg-arrhythmia-classifier-stm32 Public

    Real-time ECG arrhythmia classification on STM32F446RE using a 1-D CNN and ST X-CUBE-AI — trained on MIT-BIH, 98.14% accuracy, 460 KiB flash footprint

    C

  3. basys3-game-console basys3-game-console Public

    SystemVerilog

0