For our Intro to VLSI final project, my team and I designed a CMOS image sensor using Cadence tools. We built and simulated a 3T active pixel sensor, CMOS row and column shift registers, and column read-out circuits. We laid out a full 4×4 pixel array and verified our design with DRC and LVS checks. The system supports raster scanning, with all control signals driven by our custom shift registers. We fit the full design in a 2.5mm × 2.5mm layout and presented our results at Design Day.

This project showcases the functional verification and live demonstration of a custom silicon chip implementing a Recurrent Spiking Neural Network (RSNN)—a bio-inspired architecture designed entirely through natural language prompts to a Large Language Model (LLM), specifically ChatGPT-4.
The RSNN was synthesized from Verilog code generated by ChatGPT, validated on tasks like XOR, IRIS classification, and MNIST, and ultimately fabricated through the open-source TinyTapeout ASIC shuttle using SkyWater’s 130nm process.
To verify the fabricated chip, our team developed a complete hardware-in-the-loop test framework using microcotb, a Python-based cocotb-like environment adapted for embedded systems. We also created custom MicroPython firmware for the Raspberry Pi Pico to enable direct communication with the chip for streaming neuron parameters, injecting spike inputs, load test scripts and reading output spike trains in real time.
For Design Day, we are demonstrating the chip live using the Audio MNIST dataset: spoken digits are converted into spike-encoded patterns and passed into the RSNN on-chip, allowing us to observe how the spiking network processes real-world temporal signals in real time.

Chronic pain affects between 25% to 50% of community-dwelling elders, and up to 80% of those in institutional care. The consequences go beyond discomfort—it affects mobility and mental health and often leads to social isolation, cognitive decline, and significantly higher healthcare costs. Our SmoothRelief patch aims to be more than just a new delivery route. It’s also specifically engineered for sensitive skin. Elderly populations often experience dry or fragile skin and can be prone to bed sores and irritation. We’ve infused the adhesive with aloe and glycerin to soothe and protect, and we’ve designed the material to be gentle on delicate skin, solving a common problem with traditional adhesive patches. We have also added permeation enhancers that allow for high acetaminophen delivery for effective pain relief. Acetaminophen will be released in a steady, controlled profile for 24 hours, allowing users to be pain-free for an entire day.

This project introduces the first all-in-one scar protector patch that combines microneedling, hydrocolloid healing, vitamin C delivery, and UV protection into a single, easy-to-use solution. Designed for individuals affected by long-term scarring from surgery, injury, or acne, the patch supports skin regeneration, reduces discoloration, and shields healing skin from sun damage. By integrating clinically supported therapies into a seamless patch format, our scar protector improves both aesthetic and medical outcomes, restoring skin health and boosting confidence.