Joshua Strubel | Searcharxiv

arXiv2026

Three-Phase Evaluation of AI-Assisted Software Development Life Cycle

This paper presents an exploratory evaluation of how increasing levels of AI autonomy affect software development productivity, requirement adherence, and developer cognitive workload. A team of four developers reimplemented the same full-stack web application across three sequential phases: partial AI-assisted development using GitHub Copilot, an AI-exclusive workflow using GitHub Copilot, and an AI-exclusive workflow using AWS Kiro. Evaluation metrics included development effort (hours), requirement adherence (RITM score), AI-interaction efficiency, and NASA-TLX workload measures. Across phases, higher levels of AI autonomy were associated with reduced development effort, improved requirement adherence, and lower self-reported mental workload, while developer frustration increased modestly. The AWS Kiro phase achieved the strongest overall performance on most measured dimensions, suggesting that tooling architecture may influence outcomes independently of AI autonomy level.

cs.SE↗

arXiv2023

Detecting Neural Trojans Through Merkle Trees

Deep neural networks are utilized in a growing number of industries. Much of the current literature focuses on the applications of deep neural networks without discussing the security of the network itself. One security issue facing deep neural networks is neural trojans. Through a neural trojan, a malicious actor may force the deep neural network to act in unintended ways. Several potential defenses have been proposed, but they are computationally expensive, complex, or unusable in commercial applications. We propose Merkle trees as a novel way to detect and isolate neural trojans.

cs.CR↗