When Julia Bargouti first arrived at Georgia Tech as an undergraduate, her understanding of artificial intelligence (AI) was limited to playful applications, such as generating poems and drawings. She knew AI had transformative potential but wasn't sure how to harness it beyond entertainment and simple tasks.
Tech faculty members recognized this gap in the student experience and worked to democratize AI access. Their goal was to empower students, like Bargouti, to gain hands-on experience with AI technologies, positioning them at the forefront of AI educational opportunities and enabling them to play a leading role in its development in higher education.
This vision materialized in 2024 with the launch of the AI Makerspace, a pioneering initiative in collaboration with NVIDIA and Penguin Solutions that gives students access to the computing power needed to run today’s powerful AI models. The timing was perfect, coinciding with a surge in student enthusiasm for AI and the public release of high-profile new AI tools.
"As an undergraduate, I assumed advanced AI was out of reach,” said Bargouti, a second-year electrical engineering major. "But the AI Makerspace has shown me that I can learn about AI, improve it, and make it more effective, even as a student."
With the AI Makerspace, Georgia Tech is providing students with cutting-edge AI resources and entrusting them with integrating AI into educational frameworks.
Writing in Forbes, alumna Sophia Velastegui called it, “a pioneering model for how universities can break down barriers to advanced computing.”
Now, a year after the AI Makerspace debuted, students have laid the foundation for an efficient, accessible, and ethical resource and developed plans to scale up.
Doubling the Power
The AI Makerspace is serious technology, with serious power. Its resources have almost doubled in just the past year, ensuring students have access to computing resources typically available only to researchers or tech companies.
- Phase I: 160 NVIDIA H100 GPUS
- Phase II: 144 NVIDIA H200 GPUS
- Current Total System: 304 NVIDIA H100/H200 GPUS
Second year electrical engineering student Tejaswi Manoj presenting her ECE 2806: AI First end-of-semester project in April. The class is one of the first to completely integrate the AI Makerspace into coursework.
Julia Bargouti (left) was in the inaugural cohort of ECE 2806: AI First in the Fall 2024 semester. She served as an undergraduate teaching assistant in the Spring 2025 offering.
FIRST WAVE OF AI MAKERSPACE-POWERED COURSES
Much of the AI Makerspace usage this year can be attributed to two courses designed and taught by Professor Ghassan AlRegib in the School of Computer and Electrical Engineering.
In ECE 4252/8803: FunML (Fundamentals of Machine Learning) students learn the basics of ML then use real-world data for hands-on exercises with the AI Makerspace.
A sophomore-level course, ECE 2806: AI First (AI Foundations), is open to all Georgia Tech students and teaches AI basics — data literacy, learning, decision, planning, and ethics.
Students’ final projects use the AI Makerspace to develop a generative pre-trained transformer, or GPT, a type of language model developed by OpenAI that generates human-like text based on deep learning techniques.
“The course provided an incredible opportunity to use powerful GPUs for training our models as undergraduates,” said Tejaswi Manoj, a second-year electrical engineering student. “Without the AI Makerspace, running these processes would have taken an impractically long time. The added time allowed us to view AI creatively in a way I didn’t expect.”
With the support of the John and Marilu McCarty Chair Professorship, AlRegib developed the two courses. He noted that the AI Makerspace gives students a more confident understanding of AI: “It’s not just a practical tool. It pulls the curtain back on some of the mystery of AI. Students often tell me that using such advanced computing power deepens their theoretical understanding of the algorithms and math powering it.”
In the past year, students also have used the AI Makerspace to broaden the scope of hackathons, enhance biomedical projects, and explore AI in topics like digital twins, robotics, and art.
(text and background only visible when logged in)
(text and background only visible when logged in)
AI Makerspace in Action
As shown below, the AI Makerspace's vast computing power allows students to train AI models at incredibly fast speeds. On the left, you can see a model being run on the AI Makerspace, while on the right, it is being run on a standard on-demand cloud computing service. Before the AI Makerspace, most students, especially undergraduates, could not gain practical experience with advanced AI systems due to limited access, high costs, and insufficient training opportunities. Video courtesy of the OLIVES @ GaTech team: S. Kim, K. Kokilpersuad, G. AlRegib.
(text and background only visible when logged in)
(text and background only visible when logged in)
ENHANCING THE STUDENT EXPERIENCE
The AI Makerspace Nexus Vertically Integrated Project team has been developing applications to infuse the power of AI across the Georgia Tech educational experience.
“Technology should never get in the way of what you want to do,” said Obinna Nwachukwu, a third-year computer science student working on some of the applications. “One of the best ways to democratize the AI Makerspace is to have people unaware when they’re using it.”
The team, advised by AlRegib, has created several tools so far:
-
Study Buddy, an AI model integrated with ChatGPT and designed to assist students with their studies.
-
Timely, a scheduling tool that suggests optimal study times based on users’ schedules and preferences, helping students balance academic commitments with their personal lives.
-
A virtual assistant to streamline academic group chats and enhance efficiency in communication and study processes.
-
A mapping project using AI-driven tools to identify landmarks on campus and provide historical context and descriptions based on user-uploaded photos.
“Students often struggle with time management, leading to stress,” said Caleb Simmons, a second-year computer engineering major on the team. “We want to create applications to ease that burden.”
A selfie of the AI Makerspace Nexus taken by Professor AlRegib. The team is developing applications, tools, services, and interfaces to ensure students of all levels and disciplines can access and use the AI Makerspace, as well as to integrate it within the classroom. ECE research engineer Greg Krudysz (far back, right of the screen) is on the governance board and has been instrumental in deploying in-house developed applications and tools on the AI Makerspace.
BUILDING OUT GOVERNANCE MODEL
Every other makerspace on campus functions as an interdisciplinary resource for all the Georgia Tech community. Students have been working to organize the AI Makerspace the same way.
The ad-hoc student committee of the AI Makerspace meeting with the deans of Georgia Tech colleges in April. The ad-hoc committee has worked to formalize the organizational structure of the AI Makerspace.
Shortly after the launch, Professor Matthieu Bloch, associate dean for academic affairs in the College of Engineering, formed a committee of student representatives from other makerspaces to formalize the organizational structure of the AI Makerspace. The team recently presented their governing structure proposal to the deans of Georgia Tech’s colleges.
“Our aim is to help make the AI Makerspace accessible to everyone, with no barriers to entry,” said Rahul Gupta, a mechanical engineering major on the ad-hoc committee. “All disciplines will be transformed by AI operations, so everyone needs to engage with it.”
The AI Makerspace will implement clear rules on who, when, and how access will be granted. Students will be required to complete onboarding and training, ensuring they are well-prepared to engage with AI technologies responsibly and effectively.
“It's been a thrilling experience to see the students embrace this new venture,” Bloch said. “Their seriousness, particularly regarding the ethical components, has exceeded all expectations."
IN THE HANDS OF STUDENTS
Those newly developed tools and the student-designed organizational structure will be rolled out in Fall 2025 along with more courses that incorporate the AI Makerspace’s resources.
“Like every new opportunity, there is a mix of excitement and nerves about how things will continue to unfold,” AlRegib said. “We started out with few expectations because something like this has never been done before and we didn't know what we're capable of. But our students have answered the call.”
Simmons from the AI Makerspace Nexus team said that’s exactly the way it should be.
“For the AI Makerspace to truly transform campus, students should be expected to lead,” he said. “We have the important perspective of being on the ground, living life at Georgia Tech. It's important for us to take advantage of it and show how AI can work for everyone.”
(text and background only visible when logged in)
Related Content
Georgia Tech Unveils New AI Makerspace in Collaboration with NVIDIA
By giving students access to powerful supercomputers, Georgia Tech will teach AI to undergraduates in a way unlike any other university in the nation.
AI for Engineering
AI for Engineering aims to ensure that graduates are well-versed in artificial intelligence principles and ready to enter an AI-native workforce.
(text and background only visible when logged in)