Dependable neuromorphic computing
Testing and reliability methods for spiking neural networks, with emphasis on fault models, temporal behavior, robustness, and resource-efficient validation.
PhD researcher · Rochester Institute of Technology
Building AI systems that remain reliable when the real world gets messy.
I develop testing, evaluation, and fault-tolerance methods for intelligent systems—from neuromorphic hardware and spiking neural networks to advanced AI control.
Research focusDependable AI systems
Neuromorphic computing · AI control · Evaluations
Currently
Research Assistant, Brain LabElectrical & Computer Engineering · RIT · Rochester, New York
PhD expected 2028Research agenda
One question across multiple layers
How can we test, monitor, and recover intelligent systems before faults become failures?
Continual and adaptive SNNs
Developing resilient neural systems that can continue learning reliably in changing and imperfect operating environments.
AI control and evaluations
Studying how advanced AI systems can be evaluated and governed for dependable behavior in agentic environments.
Featured publication
Timing Matters
2026
GLSVLSI ’26 · Great Lakes Symposium on VLSI
Timing Matters: Delay Fault Characterization and Testing in SNN Accelerators
A systematic study of how axonal, dendritic, and refractory delay faults affect spiking neural networks—and how temporal faults can be characterized for dependable neuromorphic systems.
Read on ACM Digital Library DOI 10.1145/3787109.3815284
Selected projects
Research in progress
Work that extends dependable-computing ideas into learning systems and AI safety.
Resource-optimized testing of SNNs
Lightweight test generation and fault-injection methods designed to reduce the compute and memory cost of reliability evaluation while retaining sensitivity to neuron- and synapse-level faults.
Fault-aware continual learning
Improving the reliability of spiking neural networks that learn continually under hardware faults and changing conditions.
Practical AI control evaluation
Evaluating approaches for maintaining dependable oversight and intervention as AI systems become more capable and autonomous.
Experience
Research depth, engineering range
Research Assistant · Brain Lab, RIT
Testing and reliability of spiking neural networks and neuromorphic systems.
Research Assistant · Design & Intelligence Lab, Georgia Tech
Machine learning and cognitive-agent research within the NSF AI-ALOE Institute.
Software and product engineering
Built and led production software systems across education, analytics, and full-stack product development.
Contact
Let’s discuss dependable AI.
I welcome conversations about research collaboration, internships, and applied AI assurance.