Doctor of Philosophy (PhD), Computer Engineering
Expected Dec 2028Arizona State University · Tempe, Arizona · GPA 3.64
Coursework: VLSI Design, Advanced Computer Architecture, CMOS Digital Systems & Circuits, and Semiconductor Device Physics.
Profile
PhD Student & Graduate Researcher
Electrical and Computer Engineering · Arizona State University
Tempe, Arizona, United States
I am a Ph.D. student in the School of Electrical, Computer and Energy Engineering at Arizona State University and a member of the Performance and Resource Optimization in Networks (PROTON Lab), advised by Dr. Eirini Eleni Tsiropoulou. My research combines embedded firmware, wireless systems, integrated sensing and communications, data analysis, and system-level design to develop efficient hardware and software solutions.
Academic background
Arizona State University · Tempe, Arizona · GPA 3.64
Coursework: VLSI Design, Advanced Computer Architecture, CMOS Digital Systems & Circuits, and Semiconductor Device Physics.
Arizona State University · Tempe, Arizona · GPA 3.57
Professional background
Arizona State University · PROTON Lab
Western Semiconductor
Western Semiconductor
Arizona State University
Cognizant Technology Solutions
Academic community
IEEE CAMAD 2026
Selected work
Conference paper · To appear
Recognition
Physical Design
Engineering work
Trained, quantized, and deployed a six-class TinyResNet steel surface-defect classifier on an nRF5340 node. The 14,270-parameter model uses approximately 13.8 KB of quantized weights and 25.6 KB of static activation memory, achieving 97.4% test accuracy and 99.6% agreement with the floating-point model on 270 held-out images.
Developed Zephyr RTOS firmware for Nordic nRF5340/nRF7002 nodes using Wi-Fi, TCP/UDP, and MQTT. Implemented an on-device Q-learning policy that selects packet count and transmission phase using packet reliability, Age of Information, energy availability, and channel load.
Built a reproducible analysis workflow for classifying heterogeneous IoT devices and evaluating reward, energy, Age of Information, packet reliability, and transmission behavior across learning epochs, with independent validation from raw Q-matrices and device logs.
Evaluated a proposed Q-learning policy against fixed 15-packet, fixed 30-packet, and random-transmission baselines across 20-, 40-, 80-, and 100-device configurations using reliability, data freshness, fairness, energy, penalty, and channel-load metrics.
Designed a Verilog hardware accelerator for graph convolution, completed RTL-to-GDS flow, verified DRC/LVS, and used pipelining to reach 300 μm² area and 390 μW power.
Implemented and verified a 2-bit adder, synthesized the RTL, analyzed power and area, and completed automatic place-and-route in Cadence Innovus to generate GDSII.
Developed NAND and inverter layouts from gate level through physical implementation, including simulation, DRC, LVS, debugging, and parasitic extraction.
Technical foundation
7 nm, 12 nm, 15 nm, 16 nm, 32 nm, Python, Embedded C, Verilog, LabVIEW, Tcl, MATLAB
Zephyr RTOS, nRF5340/nRF7002, Wi-Fi, TCP/UDP, MQTT, embedded firmware, reinforcement learning, energy-aware transmission scheduling, QoS analysis
Synthesis, DFT, STA, CMOS, CDC, lint, floorplanning, PnR, CTS, low-power design, power and signoff analysis
Cadence Genus, Innovus, Virtuoso, ModelSim, Calibre, Xilinx Vivado, Linux, Windows, GTKWave