Inspiration

The inspiration for Explain My Day came from a common experience: finishing a day feeling tired, stressed, or unsatisfied without being able to clearly explain why. Tasks and events are easy to remember, but the real causes of mental fatigue are often subtle — poor transitions, cognitive overload, or lack of recovery. We wanted to explore how advanced AI reasoning could help people understand their days, not by tracking behavior or giving advice, but by reflecting experience back with clarity.

What it does

Explain My Day allows users to describe their day freely in unstructured text. Using Gemini 3, the application analyzes the day as a whole system and produces four structured outputs: an intelligent summary, detected friction points, hidden patterns the user may not have noticed, and three small, realistic micro-improvements for the next day. The focus is on understanding root causes rather than optimizing productivity.

How we built it

The project was built using Google AI Studio with the Gemini 3 API as the core reasoning engine. A carefully designed prompt enforces deep reasoning and strict JSON output. The frontend parses this output and renders it into a clean, single-page interface with four insight cards. The app requires no backend, no database, and no user login.

Challenges we ran into

One of the main challenges was preventing generic or motivational responses. This required precise prompt engineering to ensure the model inferred hidden patterns instead of summarizing or coaching. Another challenge was enforcing reliable JSON output while keeping the experience fast and intuitive.

Accomplishments that we're proud of

We are proud of creating a reasoning-first application that feels human, simple, and immediately useful. The project demonstrates how Gemini 3 can transform messy, real-world input into structured insight with very low latency and minimal user effort.

What we learned

We learned that simplicity amplifies reasoning. By reducing the interface to a single input and a single action, the model’s reasoning becomes the centerpiece. We also learned how powerful inference can be when AI is applied to understanding rather than generation.

What's next for Explain My Day

Future versions could include different levels of reasoning depth, trend analysis across multiple days, and applications in education, wellbeing, and workplace reflection — while preserving the core principle of clarity without complexity.

Built With

• and-json-parsing-html-?-application-structure-css-?-modern
• api-calls
• css
• gemini3
• google-studio
• html
• javascript