ClearFracture.ai is building software at the intersection of AI, data curation, data engineering, and formal systems modeling. It is a software startup developing an AI agent platform that helps automate the manual labor that is often quoted as being 80% of the work behind data analysis - data preparation. The company’s broader thesis is that AI systems need better prepared knowledge representations in order to reason effectively, and that formal systems engineering methods offer a promising foundation for representing how complex systems are structured and behave. 

The company recently released an open-source project that analyzes code repositories and generates an initial SysML v2 system model. The current version can identify and model the components of a codebase, effectively producing a component or block-diagram-style representation of the system. Brian sees this as an ideal student project because it combines open-source contribution, agentic AI, software engineering, systems engineering, and product development in a way that gives students visible, resume-relevant technical experience. 

Bowie State computer science students will extend the existing open-source AI agent by adding new functionality that moves beyond component identification into behavioral system modeling. While the current agent helps identify the parts of a software system, a richer system model should also represent what the system does, how components interact, what activities the system supports, what use cases are enabled, and what sequences occur during execution. Students will explore how AI agents or specialized agent skills can analyze code repositories and generate additional SysML v2 artifacts that describe system behavior, activities, use cases, or sequence flows. 

The project should be scoped as an applied AI software development capstone, not a theoretical research project. Students will begin by studying the existing open-source repository, understanding the current agent design pattern, learning the basics of SysML v2, and identifying one specific artifact type to generate as an extension. Potential directions include generating activity diagrams, use case representations, sequence descriptions, dependency maps, or behavioral summaries that complement the existing component model. The team should select a narrow enough feature area to prototype, test, document, and contribute back to the open-source project within the semester.

This project is ideal for students as it offers a unique opportunity to apply their skills in a real-world context, contributing to a real emerging product area with practical business and engineering relevance, working with a dynamic and forward-thinking company. Students will gain practical experience in AI and machine learning, product design and development, and UX/UI design. They will also develop critical skills such as problem-solving, teamwork, and communication, which are essential for success in the technology industry. Exploring this strategic challenge is meaningful for students as it connects academic learning with real-world career skills, providing them with a competitive edge in the job market.