Jeroglifico AI

• 

• 

• 

  code

Inspiration

We wanted to build an AI assistant that feels truly useful for students working on STEM problems from screenshots, handwritten notes, diagrams, or typed questions. Most tools either handle text well or images poorly, but not both in a clean, organized workflow. Jeroglífico AI was designed to bridge that gap with a professional multimodal interface.

What it does

Jeroglífico AI lets users upload images, ask questions in natural language, and keep separate sessions organized in a sidebar. It is built to support step-by-step explanations, context-aware reasoning, and a clean experience that feels like a real AI workspace rather than a basic demo.

How we built it

We built the front end in Streamlit and focused heavily on interface design, session management, and multimodal interaction. We implemented a custom visual style with a dark professional theme, animated background elements, sidebar-based session organization, and a chat-first workflow. The app structure was refined to keep the layout clear, compact, and easy to use during a live demo.

Challenges we ran into

One of the biggest challenges was making the interface look polished instead of cluttered. We had to reorganize the layout so the structure would not feel cut off or broken, and we also adjusted the upload flow so it fit naturally into the chat experience. Another challenge was preserving a consistent visual identity while keeping the app simple enough to use quickly.

Accomplishments that we're proud of

We are proud of creating an interface that feels much more like a product than a prototype. The app now has a cleaner structure, better organization of sessions, a more elegant visual style, and a workflow that makes multimodal problem solving feel intuitive.

What we learned

We learned that good AI products are not only about model quality, but also about how clearly the user can interact with the system. Small decisions in spacing, hierarchy, and input placement have a huge impact on whether an app feels professional. We also learned how important it is to design for a demo experience from the beginning.

What's next for Jeroglífico AI

Next, we want to connect the chat to a real reasoning backend, improve image understanding, and add richer tools for solving math and science problems. We also want to support better session saving, more interactive explanations, and a more advanced multimodal assistant experience.

Built With

• ai
• python
• vs-code