​

​

I study human-AI collaboration in software engineering, with a focus on how AI is changing the way software is specified, designed, built, and evaluated. Generative AI can produce requirements, code, tests, design recommendations, and documentation that appear polished and technically plausible, but may be incomplete, inconsistent, or simply wrong. My research examines how software teams can identify these failures before they become embedded in a system.

​

I am particularly interested in how software engineering workflows must adapt as AI-generated artefacts become part of everyday development. Existing practices often assume that artefacts are created by people who can explain their reasoning and be held accountable for their decisions. AI disrupts this assumption. Teams need stronger approaches to review, traceability, verification, and quality assurance so that generated outputs can be connected to stakeholder needs, system constraints, and prior technical decisions.

​

My work also examines how AI affects decision-making and collaboration within software teams. AI may help developers explore a problem or communicate an idea, but it may also reinforce an early assumption, make weak reasoning appear authoritative, or encourage a team to accept an output without understanding its consequences. Building on my research on perspective-taking, psychological safety, and conflict, I study whether AI helps teams make better technical decisions or simply helps them reach decisions faster.

​

These questions are especially important in requirements engineering, where business needs are translated into technical commitments. Errors introduced at this stage can shape the entire system and create significant rework later. My research investigates how teams can evaluate AI-generated requirements and designs, expose unsupported assumptions, and preserve the reasoning that connects a software artefact to the problem it is intended to solve.

​

This work also extends to software engineering education. As students increasingly use AI throughout the development process, they must learn how to assess generated artefacts, justify technical choices, and remain accountable for the systems they build. My goal is to help develop software engineering workflows and learning environments that allow teams and organizations to benefit from AI without weakening technical judgment, software quality, or responsibility.