Inspiration
Students often wait too long for answers to common assignment, exam, and scheduling questions posted online. Teaching Assistants spend excessive time repeating responses instead of focusing on higher-level guidance.
What it does
TAi is a smart assistant that automatically responds to frequently asked questions by connecting to course resources such as the syllabus, assignments, current posts, and past classes. For ambiguous questions, the AI suggests an answer for TA review, while complex cases are escalated directly to TAs.
How I built it
We designed an AI Agent Database that integrates with online Q&A platforms like Piazza or LMS systems. The system is hosted internally on university servers for full compliance with data privacy policies. The model references authoritative sources such as the syllabus and assignment pages to ensure transparency and reliability in responses.
Challenges I ran into
Ensuring ethical considerations was key — maintaining transparency in responses, building human-in-the-loop review, and aligning with strict university data privacy standards were difficult design challenges. Another challenge was balancing automation with the need for TAs to handle nuanced and complex student concerns.
Accomplishments that I'm proud of
Created a system that meaningfully reduces TA workload while giving students immediate, reference-backed support. Developed a scalable revenue model with tiered subscriptions and LMS licensing. Positioned the platform to give universities and online learning systems a competitive edge by offering AI-enhanced features.
What I learned
AI delegation can increase student self-efficacy and accelerate progress. Implementation examples in other contexts show that automation can reduce errors and dramatically improve efficiency. Research and experimentation confirmed the importance of combining automation with human oversight in educational contexts.
What's next for TAi
Expand integration beyond Piazza into other LMS platforms such as Canvas, Blackboard, and Moodle. Pilot with university courses to measure student learning outcomes and TA time savings. Refine the AI model to handle broader types of student questions while maintaining transparency and ethical safeguards.
Log in or sign up for Devpost to join the conversation.