Past Presentation | Software Excellence Alliance

The August 2026 Tech Talk was presented by Jennifer Ambrecht and Wendy Chang

Join the Discussion

Abstract

While the engineering community is increasingly familiar with what Digital Systems Engineering (DSE) is, the true challenge often lies in execution, adoption, and cultural shift. This tech talk moves beyond the foundational definitions to explore the strategic imperatives and practical methodologies for implementing DSE in today’s demanding environment.

During this session, we will discuss exactly why it is crucial for our workforce to learn and apply DSE principles to modernize our capabilities, and why DSE is the necessary future of system development. Furthermore, we will delve into the how: outlining effective learning pathways for individuals and providing actionable strategies to help existing and future projects successfully transition to DSE frameworks. Join us to gain insight into how NAVAIR/NAWCWD is actively accelerating this digital transformation to meet future mission requirements.

How AI Is Impacting MBSE: From Documents to Prompts, Agents, and Pipelines

The July 2026 Tech Talk was presented by Will Walker

Join the Discussion

Abstract

Model-Based Systems Engineering has spent a decade fighting the same battles: knowledge trapped in documents, modeling done by hand from a blank page, context siloed across a dozen tools, and a dependence on scarce experts. Artificial intelligence is now reshaping every one of those fronts and faster than most teams realize.

This talk is a grounded tour of how AI is genuinely impacting MBSE today, drawn from real, presented case studies by Ford, Dassault Systèmes, Capgemini, Quest Defense, and others. We’ll walk through six concrete impact areas: AI generating valid SysML v2 models from a single prompt; AI extracting formal, queryable models from legacy sketches and documents; automated failure and safety analysis (DFMEA) with reported 60–80% effort reductions; and of particular interest to a software audience, treating models like code, with CI/CD pipelines, the Model Context Protocol (MCP), and agentic “sense–decide–act” workflows driving the modeling tools themselves.

Just as importantly, we’ll separate signal from hype. The teams succeeding with AI all do the same unglamorous things first: they standardize with style guides and reuse libraries, and they keep a human in the verification loop. Attendees will leave with a clear mental map of where AI fits in MBSE, an honest view of the risks, hallucination, skill atrophy, IP exposure, and vendor over-promising — and a practical, five-step roadmap for adopting AI in their own practice starting Monday.

Key takeaways:

The center of gravity for AI in MBSE is generation and extraction, turning prompts, sketches, and legacy documents into valid, connected models.
Standardize before you automate: style guides and reuse libraries are what make AI output trustworthy.
Treat models like code, CI/CD and MCP turn one-off generation into continuous, agent-driven workflows.
AI augments engineering judgment; it doesn’t replace it, human verification is non-negotiable.

About the Presenter

Will Walker is the founder of CameoMagic, where he helps organizations adopt Model-Based Systems Engineering (MBSE), SysML, Digital Engineering, and AI-enabled engineering practices. Through consulting, training, and online educational content, he works with engineers across government and industry to improve system development workflows and model-based practices. As the creator of the CameoMagic YouTube Channel, he regularly shares tutorials, demonstrations, and insights on Cameo/CATIA Magic, SysML, AI agents, automation, and the future of digital engineering.

Why Battle-Tested Software Engineering Still Matters in the Age of AI

The June 2026 Tech Talk was presented by Mario Benitez

Join the Discussion

Abstract

While use of AI in developing and maintaining software is increasing globally, this use is relatively new, and proven practices for using AI effectively are still emerging. There are different ways to use AI, but their effectiveness varies with use cases, lifecycle phases, and practice variations. Activities like prototyping can move incredibly quickly, but product quality does not come for free and sharp increases in technical debt can result if software engineers do not use AI responsibly.

Software engineering principles and practices can provide significant value in guiding the use of AI in development. This talk shares observations and lessons from both personal use of AI and study of its application across multiple projects. The talk will explore different modes of AI usage from code assistants to agentic workflows, and the gap between “vibe coding” and disciplined software engineering. Finally, it will examine how AI is changing the development equation by increasing pressure as code is produced in larger volumes and at a faster pace than teams can validate and integrate incrementally.

About the Presenter

Mario Benitez is a software architect at Carnegie Mellon University’s Software Engineering Institute (SEI), where he works across Department of Defense and other government agency initiatives, from architecting large-scale systems to advancing software architecture practices. With over 20 years of experience in software engineering, he has built and delivered complex, high-reliability systems, including those in safety-critical environments. Prior to joining SEI, he spent 15 years in the railway industry, progressing from tester to product owner while leading the development of globally-deployed systems. He currently focuses on the practical application of artificial intelligence to enhance large-scale software modernization, enabling organizations to transform complex systems more efficiently and effectively.

The Data Translator in the Age of AI: Why the Most Valuable Person on Your Team Isn’t Always Writing Code

The May 2026 Tech Talk was presented by Zackary Downey

Join the Discussion

Abstract

As AI handles more of the day-to-day mechanics of data work, the bottleneck is no longer technical execution. It’s translation.

The most valuable person on a modern data team isn’t the one writing the most sophisticated code, it’s the one who can bridge the gap between messy business reality and precise, actionable direction. In the AI era, that skill is more critical than ever: getting useful output from an LLM demands the same ability as getting useful output from a data team, turning ambiguous problems into something a system can act on.

We’ll explore what the translator role looks like in practice and how leaders can identify and develop this capacity on their teams.

About the Presenter

Zackary Downey is a Data Analytics Manager at Doximity, where he leads a data analytics team in support of one of the leading platforms for medical professionals in the United States. With more than a decade of experience spanning data science, analytics, and team leadership, Zack has built and led high-performing data teams across a range of organizations: from an early-stage startup to a Fortune 5 company to a publicly traded tech company.

Before joining Doximity, Zack served as a Senior Manager of Data Science at CVS Health, where he led a team that focused on transforming the pharmacy benefits management ecosystem through predictive modeling and forecasting. Prior to that, he was Director of Data Science at Ursa Space Systems, overseeing the development of satellite image processing tools, machine learning models, and new data products derived from alternative data sources.

Zack holds a Master’s degree in Applied Statistics and Data Science from Cornell University and a Bachelor’s degree in Mathematics and Economics from Boston College. His work has consistently centered on bridging technical teams and business stakeholders, and he is passionate about people management, mentorship, and building data organizations that are both rigorous and human-centered. He recently spoke at the Software Excellence Alliance on “Empathy in AI: Improving Data Science with Kindness and Human Focus.”

The SEA Data Warehouse

The April 2026 Tech Talk announced the availability of the SEA Data Warehouse

Read the Tech Report

Warehouse Schema Overview

Download the Data

Join the Discussion

Abstract

The Data Warehouse Working Group was initiated in September 2022, with the purpose of advancing the state of the practice and the art of software engineering by providing a source of high-quality empirical data for research and benchmarking, including:

Creating a global warehouse of sanitized and aggregated data
Distributing the data and helping people to consume it
Publicizing and getting people interested in using the data
Accepting ongoing data submissions and adding them to the warehouse

This presentation will announce the availability of that data. Our intent is to encourage people to use this data to improve the state of the practice. This includes areas such as benchmarking, planning, and research. This data has been carefully organized to answer a broad set of questions needed in all areas of software development. Because the data is so well defined, the data can be reused for additional questions with high confidence.

About the Presenters

Jeff Schwalb is a computer scientist and has been supporting Naval Air Systems Command (NAVAIR) since 1984. He has over 20 years of experience developing and acquiring real-time embedded software systems for avionics, weapons, and range instrumentation systems. He also began collaborating with the Software Engineering Institute (SEI), learning and applying CMM key practices, becoming a certified Personal Software Process℠ (PSP℠) instructor and then a certified Team Software Process℠ (TSP℠) coach. Over the last 25 years he has taught and consulted hundreds of scientists and engineers in various forms of personal engineering processes and coached dozens of projects in the launch and operations of team project planning and tracking.

Dr. Bill Nichols is a Principal Engineer in the Software Solutions Division of the Software Engineering Institute at Carnegie Mellon University, where he leads the Software Measurement and Analysis team. Before joining the SEI, Dr. Nichols earned a doctorate in physics from Carnegie Mellon University, after completing graduate work in particle physics. He later led a software development team at the Bettis Laboratory near Pittsburgh, Pennsylvania, where he developed and maintained nuclear engineering and scientific software for 14 years. He has more than 30 years of technical and management experience in the software engineering industry and has published in Nuclear Instruments and Methods, IEEE Transactions on Nuclear Science, IEEE Computer, and Physical Review Letters.

Julia Mullaney has been involved with software excellence since starting her career at IBM in 1988. At IBM, she was instrumental in process improvement efforts, having a major role in defining and implementing defect prevention, for which she won the IBM Quality Award. At the Software Engineering Institute (SEI), Julia was a key contributor to the PSP℠ and TSP℠ through the development of training, certification, licensing, and applied research at leading software organizations. She was fortunate to take the first PSP class from her mentor, Watts Humphrey. Julia was chair for the IEEE CS/SEI Watts Humphrey Quality Award and sat on the IEEE CS Awards Committee.

David Tuma began his software career in the United States Air Force in 1994, where he was introduced to the high-maturity agile concepts in the Team Software Process. Impressed with TSP’s power and flexibility, he created (and continues to evolve) an open-source TSP toolset called the Process Dashboard, which has been used by tens of thousands of developers worldwide. He developed the Team Process Data Warehouse, upon which the SEA Data Warehouse was built. David values the collaboration with other members of the SEA as they seek opportunities to change the world of software engineering through broader adoption of high-maturity agile techniques.

Dr. Brad Hodgins is a computer scientist and has been supporting Naval Air Systems Command (NAVAIR) for 36 years. He has over 20 years’ experience developing simulation and avionics software. He has spent the last 16 years as a project planning and tracking coach and instructor for the Performance Resource Team (PRT), actively coaching project teams in the development of high-quality products for on-time, on-budget delivery to the fleet.

Robert Bentall is a Software Delivery Manager at Leonardo in Luton, England. Previously, he was a Principal Software Engineer at Martin-Baker Aircraft Company, and a Senior Software Engineer at Schlumberger Oilfield UK PLC.

Agentic Coding: The Earthquake That Topples the Software Factory

The March 2026 Tech Talk was presented by Larry Maccherone