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.

Glioblastoma multiforme (GBM) is the most common malignant brain tumor and remains highly lethal due to its invasive nature and resistance to treatment. Its development of oxygen-poor regions reduces the effectiveness of radiation, limiting current therapeutic outcomes. We propose a new treatment strategy that delivers oxygen to tumor cells to enhance their sensitivity to radiation and improve patient prognosis.

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.

Transitioning to renewable power generation is often difficult for remote or isolated communities due to generation intermittency and high cost barriers. Our paper presents a simulation-based optimization approach for the design of policy incentives and planning of microgrids with renewable energy sources, targeting isolated communities. We propose a novel framework that integrates stochastic simulation to account for weather uncertainty and system availability while optimizing microgrid configurations and policy incentives. Utilizing the mixed-variable Simultaneous Perturbation Stochastic Approximation (MSPSA) algorithm, our method demonstrates a significant reduction in Net Present Cost (NPC) for microgrids, achieving a 68.1% reduction in total costs in a case study conducted on Popova Island. The
results indicate the effectiveness of our approach in enhancing the economic viability of microgrids while promoting cleaner energy solutions. Future research directions include refining uncertainty models and exploring applications in grid-connected microgrids.