At ibl.ai, we're passionate about developing AI solutions that make a real difference in education. Recently, we had the privilege of collaborating with Dr. Lorena A. Barba at George Washington University's School of Engineering & Applied Science to bring a cutting-edge, personalized AI mentor to her students. This project represents a leap forward in student support, offering a 24/7 virtual version of Dr. Barba herself—ready to answer questions, guide students through coursework, and enhance their learning experience. This is what Lorena had to say about our work together in her own words:

“ibl.ai’s generative AI in education is set apart by their unparalleled expertise and their commitment to empowering educators by putting full control in their hands. Their platform enables instructors to customize AI Mentors, ensuring that the AI responses are always grounded in course materials - minimizing the risk of inaccuracies or hallucinations. ibl.ai has designed a vendor-agnostic solution, allowing us to choose the best AI model for our needs, whether it's OpenA or open models like LLaMA. This flexibility, paired with their incredible support structure, ensures that the platform stays relevant and adaptable as AI technology evolves.”

A Pedagogical Approach to AI Mentorship

What makes this AI mentor unique is its pedagogical design, tailored specifically for Dr. Barba's courses. Instead of a one-size-fits-all enterprise solution, the AI mentor offers a customizable and course-specific tool that integrates deeply with course materials. It provides context-aware responses using Retrieval-Augmented Generation (RAG) and gives students more accurate, insightful help that’s directly tied to the coursework they’re studying.

Dr. Barba put it best during a recent presentation, where she emphasized the value of focusing on cost-effectiveness and customization:

“What sets our approach apart is its focus on pedagogical design and cost-effectiveness. Unlike enterprise-wide solutions, the pay-as-you-go model could save institutions significant resources while offering greater customization.”

AWS Bedrock at the Core

Central to the development of this AI mentor was AWS Bedrock. AWS Bedrock's flexibility allowed us to build a model that wasn't limited by vendor lock-in, offering the ability to choose the best models and solutions tailored to Dr. Barba’s vision. With Bedrock, we integrated multiple AI models and fine-tuned the mentor to Dr. Barba's unique requirements—ensuring that students get a customized learning experience that evolves alongside their academic needs.

From the vendor dashboard, we tracked usage and costs in real-time, ensuring the mentor remained scalable while managing expenses. This pay-as-you-go model has made it a more accessible option for educational institutions compared to more rigid, costly alternatives.

Key Features of the AI Mentor

• Instructor-Controlled AI Persona: Dr. Barba controls the AI’s behavior and responses through customizable prompts, ensuring that the mentor aligns with her teaching philosophy.
• Course Resource Integration: The AI mentor pulls information directly from course materials, so students receive accurate, context-specific responses.
• Context-Aware Responses (via RAG): By integrating Retrieval-Augmented Generation (RAG), the AI mentor provides relevant answers based on the specific course context, not generic or unreliable information.
• Flexible Model Selection: Thanks to AWS Bedrock, we weren’t confined to a single model, allowing for greater flexibility and ensuring the best possible performance.
• Cost Usage Tracking: With Bedrock’s dashboard, Dr. Barba and her team can monitor resource usage and manage costs effectively.

Transforming Student Support

Dr. Barba’s AI mentor has already had a transformative impact on her students. During a live demo, she showcased the mentor’s interface and how it handles student queries in real-time, providing targeted feedback that encourages critical thinking. Unlike general-purpose tools like ChatGPT, this mentor is course-specific and provides guidance that is directly aligned with Dr. Barba's academic goals.

According to a survey by the Digital Education Council, 86% of students already use AI in their studies, but most rely on tools that aren't grounded in their course material. Dr. Barba's AI mentor changes that dynamic by bringing a personalized, academic-specific AI tool into the classroom—helping students leverage AI effectively while staying engaged with their course content.

The Future of AI in Education

Our work with Dr. Barba is just the beginning. We believe this model of AI mentorship, powered by AWS Bedrock, can revolutionize student support at educational institutions worldwide. By offering customizable, context-aware AI tools, educators can provide their students with an unparalleled learning experience—anytime, anywhere.

If you’re interested in how ibl.ai can help your institution develop customized AI solutions, contact us to learn more.

‍