Phase Change Scientists Present Natural Language Chat Interface Paper at AAAI Conference

Research Scientist Aleksander Chakarov, Ph.D., presented a recently published Phase Change workshop paper at the 32nd AAAI Conference on Artificial Intelligence in February. The AAAI conference, held annually by the Association for the Advancement of Artificial Intelligence, promotes research in artificial intelligence and scientific discussion among researchers, practitioners, and engineers.

The Paper

The paper, titled "Towards J.A.R.V.I.S. for Software Engineering: Lessons Learned in Implementing a Natural Language Chat Interface," was co-authored by Chakarov, Rahul Pandita, and Hugolin Bergier.

Phase Change is developing a cognitive platform and AI-based collaborative agent called Mia designed to improve software development productivity and efficiency. Mia features a natural-language chat interface enabling users to quickly get started.

Key Lessons Documented

The paper outlines lessons learned during the first iteration of Mia's chat interface implementation:

The research also explores subliminal priming applications and data-driven prioritization benefits.

Presentation Details

Aleksander presented the paper on February 2 at the AAAI Workshop on NLP for Software Engineering in New Orleans, Louisiana.

Regarding the presentation experience, Chakarov noted: "People were engaged" during the Q&A session. Two commentators provided valuable feedback — one supporting the paper's findings about subliminal priming, and another suggesting simulations as an additional tactic for faster statistical approach adoption.

The natural language interface research presented at AAAI laid early groundwork for what would evolve into Phase Change's approach to making complex codebases accessible through conversational AI — a capability now central to COBOL Colleague.