Hi, I'm Andy Wei
Electrical Engineer
Building reliable, scalable electronics for AI, robotics, and chip design
About Me
EE @ UPenn interested in hardware design, robotics, computer architecture, and AI.
Skilled in designing and optimizing energy-efficient hardware and embedded systems with expertise in circuit and PCB development using Altium and LTspice. Experienced in computer architecture design and implementation using SystemVerilog and FPGA deployment. Proficient in IoT and robotics: engineered and mentored Arduino and Raspberry Pi based embedded platforms to interface sensors and actuators for real-world applications as a teaching assistant
Computer Architecture
SystemVerilog/Verilog, FPGA, Pipelining, AXI4-Lite Interface, and Hardware Verification
Embedded Systems
C, C++, Arduino, Raspberry Pi, PCB prototyping, LTSpice simulation
Robotics
CAD Tools (SolidWorks, Fusion 360), Python, Sensor Integration, Actuator Control
AI
Python, NumPy, Scikit-learn, OpenCV, TensorFlow, Matplotlib, Custom GPT
Experience
My professional journey and key accomplishments
Head Teaching Assistant
University of Pennsylvania
- Facilitated 100+ students' mastery of analog/digital circuit design and signal-processing fundamentals in labs, leveraging oscilloscopes, logic analyzers, and microcontrollers to strengthen embedded-systems principles
- Led student teams to prototype Arduino/RPi sensor-actuator systems, manage PCB bring-up, and apply validation/debugging techniques
Electrical Hardware Engineer
Penn Electric Racing
- Led MCU board design for REV11 Power Distribution Unit with STM32, CAN interface, and current sensing; designed in Altium and validated with 10+ LTspice simulations
- Developed battery simulation circuitry to emulate high-voltage pack behavior during subsystem testing, enabling validation of cell voltage monitoring, I2C data integrity, and load response without live packs
UI/UX Designer
PennLabs
- Led design sprints for student-facing Penn mobile applications, creating Figma prototypes and high-fidelity design systems that improved task completion rates by 25%
- Collaborated closely with PMs and frontend developers to implement reusable UI components in React, ensuring cross-product UX consistency and reducing frontend design debt
Founding Robotics Engineer
Noralabs
- Achieved 30 FPS real-time perception on Jetson Nano by converting YOLOv8, pose, and fall models to TensorRT in half precision, using CUDA streams, preallocated pinned memory and BLS 2 intra-process communication
- Engineered and fabricated a 4-layer Raspberry Pi CM5 carrier PCB with isolated power-management, high-speed I2C/SPI, and GPIO expansion; routed in KiCad, impedance-controlled, and DFM-verified
- Developed bare-metal STM32 firmware in C/C++ for motor drivers, BLDC control, wheel encoders, and Li-ion battery management system; utilized oscilloscope and logic analyzer for signal integrity
Software Engineering Intern
EIQDigital
- Led development of a Slack-native employee assistant using Node.js, Slack Bolt, and OpenAI API, utilizing Pinecone vector search and role-based access controls to ensure data security, reducing 95th-percentile response time
- Passed internal security review by implementing AWS Key Management Service for secret management in Lambda, DynamoDB audit logs, and CloudWatch alerts to provide end-to-end access visibility
Undergraduate Research Assistant
University of Pennsylvania – Radway Group
- Characterized CR2032 lithium coin-cell discharge behavior under pulsed and continuous loads, logging 50+ controlled experiments with precision instrumentation to quantify capacity loss and recovery effects
- Modeled battery output via non-linear regression and derivative analysis, reducing curve-fitting error by 18% and mapping the influence of drain current, duty cycle, and pulse period on effective capacity
Projects
A showcase of my technical work and engineering projects
SecondLens
Built an AI memory chat and knowledge graph for people with Alzheimer's by implementing a Next.js and React frontend, a Node backend with named entity recognition, and Supabase Postgres for storage and authentication; Delivered an event-driven ingestion and retrieval pipeline with Mentra smart glasses
Astro Party Video Game
Recreated AstroParty on an MSPM0+ by interfacing buttons, slide-potentiometer (ADC), DAC speaker, and SPI LCD, writing low-level C/ARM drivers and configuring GPIO/ADC/SPI links for real-time gameplay. Incorporated edge-triggered GPIO interrupts to drive real-time physics calculations in the game engine
Custom Boston Dynamics Spot Robot
Integrated a ROS motion stack for a 3D-printed quadruped, implementing sit/stand/angle/walk control with ROS nodes in C++/Python and control with keyboards and joysticks. Brought up LiDAR-based mapping by running rplidar_ros and hector_slam; visualized robot state and maps in RViz; maintained TF2 tree and odometry
Custom RISC-V Processor
Developed and optimized pipeline architecture for a RISCV CPU core using System Verilog with AXI4-Lite Interface, deployed on FPGA for testing and performance
Aura
Developed a computer vision-powered desktop application with an emotion recognition system using OpenCV, TensorFlow, and Matplotlib to track and respond to the user's mental and physical fatigue in real-time. Trained a custom GPT focused on mental and physical well-being
WordHack
Designed and implemented a hardware-software system with Python using a Raspberry Pi to automate letter recognition and word tracing for the mobile game Wordhunt. Developed an accurate image processing algorithm using OpenCV and Tesseract OCR
Education
My academic background and qualifications
BSE: Electrical Engineering
University of Pennsylvania, Class of 2028
Relevant Coursework:
Get In Touch
Have a project in mind or want to discuss opportunities? I'd love to hear from you!
Location
Philadelphia, PAPhone
Available on request