Research R&D

AI and MBSE for hardware design

A framework that combines AI and MBSE so makers and small manufacturers can take an idea told in natural language to an FPGA and SoC prototype, with requirements and architecture traced in the model.

It is the method behind QvIC Labs and its cycle: idea, model, design, simulation, devkit, fabrication, integration and production.

1. IEEE LAEDC · 2025 · Best PaperAI-Driven Digital Design Prototyping with MBSE for Small-Scale Makers

Cyber-resilient semiconductors with SysML v2

An approach that uses MBSE and SysML v2 as the backbone of cyber-resilient semiconductor architectures, with security requirements traced from the specification.

1. IEEE LAEDC · 2025Integrate MBSE and SysML 2.0 for Next-Generation Cyber-Resilient Semiconductor Architectures

Engineering and STEM education

Embedded-systems and mechatronics labs designed with MBSE to adapt to industry, industry-sponsored capstone documentation supported by MBSE, and automation education for the semiconductor industry grounded in cognitive neuroscience.

This line comes from years of IEEE volunteering: I led EDS-ETC, the Electron Devices Society’s STEM education program that brings engineering to schools and teachers.

1. ASEE Annual Conference · 2026Streamlining Industry-Sponsored Capstone Documentation with MBSE: Enabling Architecture Trade Studies and Early System Level Simulation
2. ASEE Annual Conference · 2024Agile Model-Based Systems Engineering Framework to Design a Laboratory Course—Case Study: An Embedded Systems Laboratory Course
3. ASEE Annual Conference · 2024Model-Based System Engineering Applied to Designing Engineering Labs to Dynamically Adapt to Industry Trends - Case in Point: The Mechatronics, Robotics and Control Lab
4. IEEE LAEDC · 2023Accelerating Engineering Education and Workforce Development in Automation & Control for the Semiconductor Industry Based on Cognitive Neuroscience

Electronics for space

My new line: resilient electronics for space, data analysis on remote hardware to estimate asteroid composition, and FPGA-based autonomy for devices on the Moon and Mars.

The starting point is a CNN that classifies asteroid taxonomy from SMASS II spectra, with 96% test accuracy on 56×56 pixel images, a small footprint meant for resource-constrained hardware.

Patents

Three patents that bring together AI, education and agribusiness: one already granted and two pending.