Ambar Murillo
Research Areas
Authored Publications
Sort By
Understanding and Designing for Trust in AI Powered Developer Tooling
Ugam Kumar
Quinn Madison
IEEE Software (2024)
Preview abstract
Trust is central to how developers engage with AI. In this article, we discuss what we learned from developers about their level of trust in AI enhanced developer tooling, and how we translated those findings into product design recommendations to support customization, and the challenges we encountered along the way.
View details
Preview abstract
The evolution of AI is a pivotal moment in history, but it’s not the first time we have experienced technological advances that have changed how humans work. By looking at the advances in automobiles, we are reminded of the importance of focusing on our developers' needs and goals.
View details
Preview abstract
AI-powered software development tooling is changing the way that developers interact with tools and write code. However, the ability for AI to truly transform software development depends on developers' level of trust in the tools. In this work, we take a mixed methods approach to measuring the factors that influence developers' trust in AI-powered code completion. We identified that familiarity with AI suggestions, quality of the suggestion, and level of expertise with the language all increased acceptance rate of AI-powered suggestions. While suggestion length and presence in a test file decreased acceptance rates. Based on these findings we propose recommendations for the design of AI-powered development tools to improve trust.
View details
Preview abstract
ML enhanced software development tooling is changing the way software engineers develop code. While the development of these tools continues to rise, studies have primarily focused on the accuracy and performance of underlying models, rather than the user experience. Understanding how engineers interact with ML enhanced tooling can help us define what successful interactions with ML based assistance look like. We therefore build upon prior research, by comparing software engineers' perceptions of two types of ML enhanced tools, (1) code completion and (2) code example suggestions. We then use our findings to inform design guidance for ML enhanced software development tooling. This research is intended to spark a growing conversation about the future of ML in software development and guide the design of developer tooling.
View details
Preview abstract
Software developers write code nearly everyday, ranging from simple straightforward tasks to challenging and creative tasks. As we have seen across domains, AI/ML based assistants are on the rise in the field of computer science. We refer to them as code generation tools or AI/ML enhanced software developing tooling; and it is changing the way developers write code. As we think about how to design and measure the impact of intelligent writing assistants, the approaches used in software engineering and the considerations unique to writing code can provide a different and complementary perspective for the workshop. In this paper, we propose a focus on two themes: (1) measuring the impact of writing assistants and (2) how code writing assistants are changing the way engineers write code. In our discussion of these topics, we outline approaches used in software engineering, considerations unique to writing code, and how the disciplines of prose writing and code writing can learn from each other. We aim to contribute to the development of a taxonomy of writing assistants that includes possible methods of measurement and considers factors unique to the domain (e.g. prose or code).
View details
"If I press delete, it's gone" - User Understanding of Online Data Deletion and Expiration
Andreas Kramm
Sebastian Schnorf
Proceedings of the Symposium on Usable Privacy and Security 2018
Preview abstract
In this paper, we present the results of an interview study with 22 participants and two focus groups with 7 data deletion experts. The studies explored understanding of online data deletion and retention, as well as expiration of user data. We used different scenarios to shed light on what parts of the deletion process users understand and what they struggle with. As one of our results, we identified two major views on how online data deletion works: UI-Based and Backend-Aware (further divided into levels of detail). Their main difference is on whether users think beyond the user interface or not. The results indicate that communicating deletion based on components such as servers or "the cloud" has potential. Furthermore, generic expiration periods do not seem to work while controllable expiration periods are preferred.
View details
