Full-Stack Hardware Engineer

Ben Levy

I work at the seams of hardware, electrical, and software. I've carried products from mechanical design and custom PCB through firmware bring-up, motion control, and automated test, in both pre-seed and regulated medical environments.

LinkedIn ↗ Email ↗ Resume ↗

Featured Work

Layer jamming mechanism with force and vacuum LAYER JAMMING F V
Research · IEEE Haptics 2024

S.E.S.H. Glove

Novel vacuum-based layer jammer wearable for VR haptic feedback. First-author IEEE publication. Designed, prototyped, and validated the full system.

4N blocking force
28:1 transparency ratio
<$0.20 per jammer
BeatRite data pipeline: sensors to BLE to iOS app to cloud BLE nRF52 SENSORS WIRELESS iOS APP CLOUD
Medical Device · Pre-Seed Startup

BeatRite

I was the entire engineering function at a pre-seed medical wearable startup. I took the prototype from breakout boards to a custom 4-layer PCB, and owned the tradeoff the company kept running into: better signal costs parts, power, and time we didn't have.

Sole technical lead
Roadmap + architecture
1st engineer hired

All Projects

Elbow exoskeleton with programmable resistance PROGRAMMABLE RESISTANCE
Mechatronics · Senior Design

BERR EXO

Portable active-resistance elbow exoskeleton built by a 4-person senior design team, which I lead. Torque varies with joint position at 50 Hz instead of holding a fixed load. Won both top awards at the UW–Madison senior design expo ($6,000 total).

Training planned from my own activity data TRAINED ON MY OWN DATA
Personal Product · iOS + watchOS

BenLift

An AI fitness tool built around how I actually train. My Apple Watch already tracks every ride, run, and climb along with sleep and HRV, so BenLift uses that to plan what I should do in the gym, adjusts during the session, and learns my tendencies over time.

Natural language to database query system ⚡ NL → SQL IN SECONDS
Enterprise AI · GE HealthCare

AI ARC OPS

Retrieval-augmented tool that turns plain-English questions into SQL. Cut workforce database query time from 45 minutes to under 10 seconds and unblocked 12+ managers' decision workflows. Won the CTO's "3 Heartbeats" Award.

24/7 automated stress testing 100K cycles 24/7 AUTOMATED STRESS TESTING
Robotics · GE HealthCare

Monte Carlo Testing Robot

24/7 simulation and test framework for anesthetic delivery equipment. 100k+ simulated breathing cycles driven through 3D-printed hardware interfaces for unattended stress and regression testing.

Automated optical testing system ML Ready PRECISION OPTICAL AUTOMATION
Imaging Hardware · Glass Imaging

Camera Prototypes & Optical Test Automation

Built a working security camera prototype from off-the-shelf hardware to validate the imaging pipeline ahead of custom optics, designed lens housings for unreleased optics, and automated the photo data collection rig — cutting experiment duration by 30%.

ML image classification achieving 98% accuracy 98% ACCURACY TRANSFER LEARNING CLASSIFICATION
Machine Learning · MSE 361

ML Fracture Classification

Binary SEM fracture classification comparing 4 architectures. ResNet50 achieved 98% accuracy, pruned to 97.8%.

Global safety incident reporting app ✓ GLOBAL SAFETY REPORTING
Full-Stack iOS

Us4Her

Solo-built Swift iOS app for global incident reporting. Frontend, backend, and database. Built in high school.

Glove-controlled robotic hand ★ WHERE IT ALL STARTED
Hardware · Origin Story

Robotic Hand

Potentiometer glove controlling a servo-driven hand. First Arduino and PCB experience. The project that started it all.

Skills Network

Core competencies across four engineering domains. Explore the interactive graph below for the full picture.

Hardware / Mechanical
SolidWorks CREO GD&T FEA CNC Machining 3D Printing
Electrical / Embedded
KiCad PCB Design SPI / I2C / UART Microcontrollers BLE Biosensors Bench Debug
Software / iOS
Python C/C++ Swift SQL MATLAB/Simulink Automated Test
ML / Data
Real-Time Signal Processing PyTorch RAG Architecture Transfer Learning Azure LLMs
Hardware/Mechanical
Electrical/Embedded
Software/iOS
ML/Data

Experience

Aug 2025 – Aug 2026

Product Engineer

BeatRite

Sole technical lead at a pre-seed medical device startup — the entire engineering function. I owned the system architecture and roadmap for a wearable remote patient monitoring platform, and designed a 4-layer custom PCB in KiCad that moved the prototype off breakout boards to surface-mount and resolved sensor front-end signal integrity faults, worked on the iOS app and BLE data path, and hired and managed our first electrical engineer. Being the only engineer meant the harder half was translating what the clinical side needed into buildable requirements, arguing signal fidelity against very limited resources, and defending those tradeoffs to investors.

Aug 2025 – Dec 2025

Applied AI Engineer (Contract)

GE HealthCare

Promoted from co-op to own internal AI tooling, using LLMs to automate company-wide budget allocation. Continued building on the AI ARC OPS platform I developed during the co-op.

Jan 2025 – Aug 2025

V&V Automation & Test Tools Engineering Co-op

GE HealthCare

Built a 24/7 automated Monte Carlo simulation and test framework for anesthetic delivery equipment, running 100k+ simulated breathing cycles through 3D-printed hardware interfaces for unattended stress and regression testing. Separately shipped "AI ARC OPS," a retrieval-augmented generation tool that cut workforce database query time from 45 minutes to under 10 seconds and unblocked 12+ managers' decision workflows — awarded the CTO's 3 Heartbeats Award.

Jun 2024 – Dec 2024

Mechatronics Engineering Intern

Glass Imaging

Built a working security camera prototype from off-the-shelf hardware to validate the imaging pipeline on real hardware ahead of custom optics. Designed and tested novel lens housings for unreleased optics, iterating mechanical concepts against the ML team's imaging requirements. Also built an automated photo data collection system, instrumenting the rig with sensors to monitor capture conditions and enforce dataset quality, which cut experiment duration by 30%.

Nov 2022 – May 2024

Undergraduate Researcher

Madison Machine Interaction Lab

Identified an unmet need in VR haptics and pioneered a novel method of using layer jammers as actuators for feedback devices. I designed, fabricated, and tested the S.E.S.H. Glove system and published the results as First Author at IEEE Haptics Symposium 2024.

Expected Dec 2026

BS Mechanical Engineering

University of Wisconsin-Madison

3.83 GPA · 5x Dean's List

Coursework: Mechatronics, Instrumentation & Measurements, Advanced CAD, Metals and Polymer Manufacturing.

Download Resume (PDF) ↗

About

Ben Levy headshot

I work at the seams of hardware, electrical, and software — the places where a project stops being anyone's clear responsibility.

I've carried products from mechanical design and custom PCB through firmware bring-up, motion control, and automated test, in both pre-seed and regulated medical environments. From a robotic hand in high school to IEEE-published haptics research to sole technical lead on a medical wearable, I'm drawn to problems that need the whole stack to solve — and to the seat where engineering decisions and business decisions are the same decision.

Currently finishing my BS in Mechanical Engineering at UW–Madison (Dec 2026). Originally from the Bay Area.

Education BS Mechanical Engineering, UW–Madison
Location Sunnyvale, CA ↔ Madison, WI
Leadership Theta Tau Xi, Exec Board. $4,000+ sponsorships secured
Languages Hebrew (native), English (native)
Outside of Work Surfing, bouldering, running, hiking, sailing, and anything outdoors