TeachAI

Inspiration

Teachers in secondary education often have to juggle lesson planning, grading, and mentoring—tasks that leave them with little time to provide personalized student support. Meanwhile, students who are still developing English proficiency can easily fall behind in a system where most educational resources are offered exclusively in English. TeachAI was conceived as a tool that would both lighten the grading and administrative burden on teachers, thereby freeing them to focus more on their students’ needs, and level the playing field for learners of all language backgrounds.

What It Does

TeachAI functions as an AI-driven teaching assistant, empowering students to ask homework and lecture-related questions on a unified platform. If a student feels more comfortable learning in another language, TeachAI provides text-based and audio explanations in over 30 different languages. This inclusive feature ensures that language proficiency is no longer a barrier to classroom success. On the teacher’s side, TeachAI helps with grading, offers a chatbot that keeps them informed on the topics students struggle with the most, and integrates email functionality to send class-wide updates. It also syncs with the Zoom API and utilizes AI summarization capabilities, enabling teachers and students to revisit concise summaries of their virtual classes whenever they need to review or catch up.

How We Built It

Our frontend was created using Next.js, where we focused on user-friendly design to make interacting with AI services intuitive for both teachers and students. The backend is powered by Flask, which communicates with our chosen AI models and manages user data seamlessly. We used Google Gemini for extracting text, summarizing Zoom meetings, and automatically grading assignments, all in an effort to reduce the burden on teachers so they can spend more time guiding their students. Meanwhile, OpenAI facilitates the core chatbot capabilities, giving teachers and students alike a helpful Q&A environment. To enhance accessibility further, we integrated ElevenLabs, which generates audio lessons and content in 30 different languages, ensuring that TeachAI can cater to diverse linguistic needs. Our team collaborated through version-controlled workflows, dividing tasks efficiently and bringing everything together through regular check-ins and merges.

Challenges We Ran Into

One significant challenge was working around the limitations of various API endpoints, which allowed only a limited number of attempts. We had to be strategic about testing and debugging to ensure we could fully utilize the AI services. As our project expanded, it also became increasingly difficult to pinpoint whether bugs were rooted in the frontend or the backend, especially since new features often involved updates on both sides. Additionally, version control grew complicated when multiple branches introduced features in parallel, leading to complex merge conflicts that required careful resolution.

Accomplishments That We’re Proud Of

We are proud to have brought together a working minimum viable product within such a short time frame, particularly given that we conceived the concept when we arrived at the hackathon. Combining different AI tools—Gemini for automated grading and meeting summaries, and OpenAI for real-time Q&A—was especially rewarding because it demonstrated how these technologies can meaningfully lower barriers for teachers and students. Our team’s ability to collaborate, coordinate, and synchronize efforts despite the usual hectic pace of hackathon development was another highlight, and it strengthened our confidence in tackling future, more complex challenges.

What We Learned

Perhaps the most valuable takeaway from this experience was the importance of strategic planning and communication within a team. Rather than jumping directly into coding, we devoted time to hashing out architecture, task delegation, and timelines, which helped us work more efficiently. This hackathon also gave us an opportunity to sharpen our full-stack skills as we integrated multiple AI services, tackled real-world use cases like grading and audio generation, and learned to optimize our testing approach under strict API usage limits.

What’s Next for TeachAI

In the future, we aim to expand our platform’s functionality by developing a comprehensive student management system that will allow teachers to track individual progress and focus on learners who need additional help. We also plan to refine our AI-driven analytics to provide deeper insights into the common areas where students struggle, enabling teachers to adjust their lesson plans proactively. Continuing to streamline our user experience will remain a priority, as we want to make TeachAI as intuitive and accessible as possible. Ultimately, our vision is to bring TeachAI to educators and students worldwide, ensuring that language limitations and excessive teacher workload never stand in the way of quality education.