This week was an exciting milestone for our IPPD team as we completed our Preliminary Design Review (PDR) and traveled to RTX headquarters in Largo, Florida, to present our work.

We spent the beginning of the week finalizing our slides, reviewing our research findings, and preparing to explain the technical direction of AuditBot. I focused on refining the executive summary, literature review, and research findings portions of the presentation to make sure we clearly communicated what we had learned and how those findings influenced our proposed design.

After completing our preparations, we visited RTX’s Largo facility, where we presented alongside another RTX-sponsored IPPD team and a team from the University of South Florida (USF). It was a great opportunity to share our progress with industry professionals and see the different engineering problems other student teams were working to solve.

We also had the opportunity to tour the RTX facility, which was one of the highlights of the week. Seeing the engineering environment firsthand gave us a better appreciation for the real-world applications of our project and the importance of developing reliable software for industry use.

One of my biggest takeaways from the experience was the importance of being able to clearly explain technical decisions and connect research findings to practical engineering requirements. Preparing and delivering the presentation also helped us better understand which parts of our proposed system need further refinement.

With our PDR now completed, we’re looking forward to incorporating feedback and moving further into the implementation phase of AuditBot. Our next steps will involve refining the architecture and beginning work on the foundational components of the system.

This week, our team focused on preparing for our upcoming Preliminary Design Review (PDR) for AuditBot, our RTX-sponsored project aimed at assisting with the NIST Risk Management Framework (RMF) compliance process using offline AI.

A major part of the week involved bringing together the research, planning, and technical discussions completed over the past several weeks. The team worked on organizing the presentation to clearly communicate the problem being addressed, our proposed solution, and the technical decisions behind our approach.

Our team continued refining the executive summary, literature review, and research findings, particularly focusing on what we learned about the RMF process and how locally hosted language models could assist with compliance-related tasks. We also worked on presenting our initial model evaluation findings and explaining the importance of offline deployment in environments with restricted network access.

Additionally, the team reviewed our proposed system architecture, including document ingestion, control mapping, evidence management, and local LLM inference. These discussions helped ensure that the different components of AuditBot were clearly defined and aligned with the project’s overall objectives.

One of the biggest challenges this week was translating our technical research into a clear and organized presentation. The team focused on communicating our findings in a way that would be understandable to both technical and nontechnical audiences while distinguishing between completed research and planned implementation.

Looking ahead, our next priority is to finalize the PDR presentation and prepare to present our progress to RTX at their Largo facility. This will be an important opportunity to receive feedback on our proposed design and identify areas for improvement as we move toward implementation.

This week, our team held our second meeting with our Raytheon liaison and continued refining our understanding of the project’s scope. We began researching NIST standards and security controls to better understand the Risk Management Framework and how it will influence our solution. We also started connecting project requirements to potential software components and development goals. Next week, we plan to define clearer team roles and begin brainstorming potential system architectures for liaison review.

This week, our team officially kicked off the Raytheon project by meeting with our liaison and gaining a clearer understanding of the project goals and expectations. We began researching offline AI models and tools such as Ollama that could potentially support our solution. We also established recurring meetings with our Raytheon liaisons, created our team name and logo, completed our team charter, and finished IPPD lab training. Next week, we plan to compare potential offline AI models and identify the best options for our use case.

This week, our team focused on getting organized and preparing for the start of our Raytheon project. We established regular meeting times, began building our team dynamic, and held an introductory meeting with our IPPD coach. We also completed lab safety training and obtained access to the IPPD lab space. Next week, we plan to create our team name and logo, meet with our Raytheon liaison to better understand the project goals and expectations, and begin learning more about training AI models.