Artificial intelligence has undeniably revolutionized the realm of software engineering, with generative AI and large language models (LLMs) capable of producing vast quantities of code and documentation. Machine learning algorithms now routinely monitor system performance and proactively identify security vulnerabilities. However, the transformative impact of these AI tools on the conception, design, and fabrication of complex physical systems, such as a jet engine, remains a critical question. This past semester, the JARVIS Challenge, an intensive sprint focused on Jet-engine AI Research and Validation, was established to investigate precisely this. The initiative tasked MIT undergraduates with discovering whether AI could significantly compress the traditional design-build-test cycle, ultimately enabling them to build faster and with greater efficacy.
"The JARVIS challenge demonstrated that AI can substantially accelerate safety-critical hardware engineering, but engineering judgment remains the decisive differentiator," stated Professor Zolti Spakovszky, director of the MIT Gas Turbine Laboratory. "An AI-native engineer is not defined by using AI, but by leading it—knowing when to trust it, when to challenge it, and how to translate AI outputs into working hardware. Manufacturing, not engineering design or analysis, remained the fundamental rate-limiting step."
The Teams, The Tools, and The Ambitious Task
The JARVIS Challenge presented a formidable four-week undertaking for undergraduates: to design, fabricate, assemble, and test a small gas turbine aero engine. The core directive was to leverage AI as their primary engineering partner throughout this rigorous process. The ultimate objective was to construct a "JARVIS-class" single-spool jet engine capable of producing between 50 and 100 pounds of thrust, running on Jet-A fuel, and successfully completing five 60-second operational runs. Teams were granted complete autonomy over their design choices, material selections, and fabrication methods.
Representing a broad spectrum of disciplines across the School of Engineering, 31 students organized into seven distinct teams. These teams comprised a diverse mix of experience levels, from all-first-year groups to those heavily populated by senior undergraduates. Many of the competitors entered the challenge with limited prior experience in specialized fields such as turbomachinery, compressible flows, or, particularly for the younger students, thermodynamics. Indeed, a significant number had never even seen the inner workings of a gas turbine before committing to building one.
To facilitate their ambitious goals, the student teams had access to MIT’s state-of-the-art machine shops and a network of trusted manufacturing vendors. They also utilized commercial software packages, including industry-standard tools like Concepts NREC for aerodynamic design, SolidWorks for 3D modeling, and ABAQUS for finite element analysis. A suite of specialized test rigs was also available for characterizing and assembling individual engine components.
Crucially, the teams were provided with access to MIT Parley, a newly launched platform designed to aggregate cutting-edge large language models through a unified interface. This provided the JARVIS leadership with direct insight into how the students were engaging with the AI tools, tracking their prompts, the associated costs, the specific LLMs being utilized, and other vital operational data. The JARVIS leadership secured early access to Parley for all participants, and with substantial financial backing from MIT Lincoln Laboratory, the Department of Mechanical Engineering, and corporate sponsors Safran, Voyager Technologies, Beehive Industries, and Boom Technology, students were afforded virtually unlimited access to AI resources.
The sponsors’ involvement was driven by a keen interest in recruitment and a genuine curiosity about the potential for AI to fundamentally reshape engineering workflows. Ryan (Hal) Hefron of Voyager Technologies emphasized this point to the students, stating, "We see this as the future of engineering. You’re honing skills that are not just nice to have—they’re going to be the future baseline in the engineering workforce."
Vincent Garnier, managing director of Safran Tech, observed the competition with keen interest. "JARVIS was a genuine experiment, a learning endeavor," he remarked. "We frankly didn’t know what to expect, from the students or from the AI models. What struck me coming from the students was: first, the enthusiasm to explore; then, as the project developed, they all came to the cool-headed realization of what AI could or could not help them with, and then almost instantly adapted for that. It makes me confident that this generation of leading engineers will probably not fall prey to easy and shortsighted use of AI, and will do so by keeping ever more in contact with experiments—physical or thought experiments."
The faculty leadership—comprising Professors Zachary Cordero, Zolti Spakovszky, Masha Folk, and Andreea Bobu from the Department of Aeronautics and Astronautics, along with engineers from Lincoln Laboratory and a dedicated team of teaching assistants—provided essential guidance and ensured the safety protocols were strictly adhered to. During weekly progress reviews, they critically assessed the students’ advancements and observed their methodologies in employing AI. Professor Spakovszky developed a nuanced approach to guiding teams, offering direction without providing direct solutions or unsolicited assistance. Following a team’s presentation, he might pose a question such as, "Do you know what a rabbet fit is? Take in the comment." This Socratic method encouraged critical thinking and self-discovery.
AI’s Dual Role: Facilitator and Hindrance
By the conclusion of the first week, one team had withdrawn from the competition. The remaining teams, with varying degrees of success, had formulated initial designs for their gas turbines. AI was employed across diverse functions: summarizing technical literature, providing instruction on design software operation, identifying potential vendors, generating comparative analyses for design decisions, answering specific technical queries, and sourcing relevant academic references. One particularly innovative team even established an AI agent within Parley to function as their project manager.
As week two commenced, teams transitioned to the intricate phase of detailed CAD design, procurement of components, and prototyping of their combustors. It was at this juncture that the limitations of AI utilization began to surface. While LLMs like Claude and ChatGPT proved adept at suggesting design alternatives and bridging knowledge gaps, the notorious characteristics of generative AI—such as hallucinations, sycophancy, and a deficit in genuine physical understanding—started to erode team confidence and decelerate progress.
"AI is a helpful tool, great at finding information, helping organize things, and can write well, but it can’t do design," commented Elizabeth Tupaj, a member of the team "811 Crew." "The moment the engineer doesn’t know what is going on and the AI is in charge is the moment the design becomes unreliable, at least with AI at its present capabilities."
Teaching assistant John Zhang observed this phenomenon firsthand, noting, "Seeing this firsthand with the students reminded me how much first impressions matter. If the students couldn’t get answers from the AI early on, they quickly grew frustrated and formed a lasting opinion that precluded them from using it later."
In the critical final weeks, the finalist teams encountered another significant obstacle that no AI solution could surmount: navigating vendor relationships. Students reported, "AI searches found vendors we had no rapport with, who had no interest in our tight timeline. The vendors who came through were the ones our team had personal relationships with." This underscored the enduring importance of human networking and established professional connections in practical engineering execution.
Among the three finalists, only the "Fast and Fractured" team achieved first-attempt ignition of their mini-combustor. This team had extensively utilized AI for trade studies and architectural comparisons, ultimately arriving at a viable design despite none of its members possessing prior gas turbine experience.
Masha Folk, the Charles Stark Draper Career Development Professor of Aeronautics and Astronautics, reflected on the project’s success: "The JARVIS Challenge showed what’s possible when you combine AI-enabled design with motivated students and a culture of rapid experimentation. The moment that stood out most was when the first student-designed combustor was installed on the test stand. It ignited flawlessly, ramped to full power, transitioned to dual-fuel operation, and then sustained stable combustion on 100 percent Jet-A fuel. This was proof that we can dramatically accelerate the cycle of design, build, and test while giving students hands-on experience with a real engineering challenge."
At the Vanguard of AI-Native Engineering
By the end of May, two of the more senior teams, "Fast and Fractured" and "811 Crew," successfully completed full engine tests. "Fast and Fractured," despite their AI-assisted design prowess, faced significant delays due to vendor issues that persisted week after week before they could reach the testing phase. Unfortunately, their hot fire test was prematurely halted when the rotor experienced friction and seized against the stationary housing. The "811 Crew," however, who entered the competition with greater prior exposure to turbomachinery and propulsion concepts, emerged as the victorious team. Their engine successfully started, transitioned to Jet-A fuel, and generated net thrust.
PhD student Joe Chiapperi described the intense atmosphere: "As we stood there with the air-starter, hearing their engines spool up and watching them spit fire, it felt like my heart was racing out of my chest. There were so many ways it could go wrong! What these students accomplished in such a short time span is nothing short of amazing."
Intriguingly, the "811 Crew" had adopted a more resistant stance towards extensive AI use throughout the competition, prioritizing their foundational knowledge and teamwork. "We had people who were at least somewhat familiar with the design software, mechanical engineers who knew how to build anything, and aerospace engineers who had taken classes on the design of gas turbine engines specifically," explained Tupaj.
An interesting observation emerged regarding the students’ engagement with AI: younger students tended to utilize Parley more frequently and with greater ingenuity, while juniors and seniors leveraged their more substantial prior experience. Professor Andreea Bobu offered a key insight: "JARVIS taught me that getting value from AI takes two things: enough expertise to judge what it tells you and catch it when it’s wrong, and enough curiosity to actually lean on it where it could help. The team that moved fastest in the sprint was experienced and leaned heavily on AI to get there. The team that eventually won was more resistant to AI; they had the expertise, but that skepticism made them slower. The sweet spot seems to be knowing enough to stay in charge of the tool, and being eager enough to pick it up in the first place. To me, that’s the real opportunity ahead: training the next generation of engineers who have the judgment to direct these AI tools and the instinct to reach for them."
The competition’s most unequivocal finding reinforced a long-held tenet of engineering: experience acts as a significant multiplier, and the human element remains indispensable. A firm grasp of first principles and fundamental concepts cultivates sound engineering judgment, which is crucial for navigating complex decision-making processes when faced with incomplete information. Moreover, when it comes to the construction of safety-critical physical systems, human oversight, hands-on execution, and ultimate accountability are irreplaceable.
"JARVIS has shown that AI copilots can have a multiplicative effect on engineering productivity, with judgment and first-principles thinking serving as the key differentiators among teams," added teaching assistant Kyle Woody.
The broader implications of AI in the aerospace sector are profound. If small, agile teams, empowered by effectively managed AI copilots, can condense design-build-test cycles from years down to mere weeks, the ramifications for workforce structure, research and development timelines, and competitive dynamics within the industry could be substantial. The students who participated in the JARVIS Challenge are among the pioneering engineers to confront these stakes not as a theoretical exercise, but within the tangible environment of a machine shop, with a jet engine humming on a test stand.
"JARVIS highlighted the power of AI in the design of physical systems," concluded Professor Cordero, associate director of the MIT Gas Turbine Laboratory. "But it also showed that the key to unlocking that power is education, through coursework, internships, and hands-on extracurriculars like MIT Motorsports and Rocket Team. Performance in JARVIS correlated strongly with year in school. My main takeaway is that in the AI era, education is more valuable than ever."
