Teach with IA

Learn with AI at your own pace: receive feedback, identify your strengths and areas for improvement, and have a real tutor who validates your progress.

Comment

Inspiration

Many students move forward without knowing exactly what they're doing wrong or how to improve. Traditional education doesn't always adapt to each student's individual needs, while learning with AI alone can lack human validation and guidance. That's why this idea was born from the need to create a more personalized learning experience, where AI adapts to each student, detects their errors and strengths, but without replacing the guidance and validation of a real tutor.

What it does

It's a system that doesn't just teach, but understands how each person learns, detects why they're struggling, adapts its strategy in real time, and allows a human tutor to validate and guide their progress. The system shows which concepts you truly master, which mistakes you repeat and why, how you learn best, which skills you're developing, what knowledge you're forgetting, what your strengths and weaknesses are, and how you evolve over time.

How we built it

We looked for problems that felt important to us, and among them we saw education and noticed how little it has evolved over the years. That's when we thought AI could take on part of the work, but since AI can also make mistakes, we wanted to bring a teacher into the loop to supervise it. We started by planning how the users (teacher and student) would communicate with the AI, and once we had a clearer idea of the workflow, we looked for technologies that would let us build faster. With the help of Google Antigravity, we were able to bring our idea to life.

Challenges we ran into

We had never built AI powered software before, we had only ever used AI through chat, so integrating AI into this project was completely new to us, as was working with RAG for the first time. This was also our first time seeing the inner workings of hackathons and startups.

Accomplishments that we're proud of

This is the first hackathon we've ever entered in our lives, and we hope it's the first of many. We always had scattered ideas but never actually followed through on them. With this project, we made more progress than we ever had before and built an MVP we're genuinely proud of, one we believe has great potential.

What we learned

As mentioned above, almost everything we did was new to us. We learned that AI isn't just a chat interface, it can be used in large systems to perform surprising tasks that save people a lot of time. We also learned about RAG and how important context is so that AI doesn't hallucinate and gives more accurate answers. And we learned that we really are capable of building the projects we plan. With this completely new experience, we hope to take part in more hackathons, build startups, and above all, create systems that can help people and generate real social value.

What's next for Teach with AI

Reaching organizations, schools, states, and governments so they can implement Teach with AI. The idea will evolve into a personal, holistic development AI that doesn't just teach content, but builds a living map of each person: what they know, what they can do, what they're missing, how they learn, what their strengths are, what they want to achieve, and what opportunities fit their profile. Their learning will continuously adapt, while tutors and experts validate their competencies through real evidence. This way, the platform will connect learning, skill development, and professional opportunities, supporting each person from who they are today to who they want to become.

Built With

Share this project:

Updates