Parallel You

• 

  Where you can ask anything

• 

  Homepage/Landing Page

• 

  What you get from Google Gemini

Inspiration Making decisions is hard not because we lack information, but because we cannot see how different ways of thinking would change the outcome. We often wonder what would happen if we were more cautious, more ambitious, or more focused on long term fulfillment. Parallel You was inspired by this gap. We wanted to create a way to explore how the same decision can lead to very different futures depending on the values and reasoning style behind it.

What it does Parallel You is a multi perspective decision simulator. Users input a decision or dilemma they are facing, along with optional context. The app then generates multiple parallel versions of the user, each operating under a different value system such as optimization, long term strategy, risk minimization, or personal fulfillment.

Each parallel version explains the choice it would make, the reasoning behind it, and the likely short term and long term outcomes. This allows users to compare tradeoffs clearly instead of relying on a single answer or intuition.

How we built it The project was built as a web application using Lovable for rapid UI development and Google Gemini as the core reasoning engine. Gemini is used to simulate multiple independent agents within a single structured prompt. Each agent follows strict constraints to maintain consistent values and reasoning across the simulation.

The frontend focuses on clarity and comparison, displaying each parallel version side by side in a structured format. The app is fully stateless and requires no authentication, allowing users to experiment freely and rerun simulations quickly.

Challenges we ran into One of the main challenges was preventing the AI outputs from blending together. Without careful prompt design, different perspectives can collapse into similar answers. We solved this by enforcing clear agent roles, explicit value priorities, and structured output requirements.

Another challenge was balancing depth with readability. We wanted the reasoning to be meaningful without overwhelming the user, which required careful formatting and hierarchy in the UI.

Accomplishments that we're proud of We are proud of creating an experience that feels thoughtful rather than gimmicky. Each parallel version maintains a distinct voice and reasoning style, making the comparisons genuinely useful. We also successfully demonstrated multi agent reasoning using a single Gemini integration, showing how the same model can produce diverse and consistent perspectives.

What we learned We learned that the power of large language models is not just in generating answers, but in structuring reasoning. By constraining values and roles, the same model can behave like multiple independent thinkers. We also learned the importance of designing AI systems that help users reflect and understand tradeoffs instead of simply giving advice.

What's next for Parallel You Future improvements include allowing users to customize value systems, save and compare past simulations, and explore more complex decision timelines. We also want to experiment with visualizing outcomes probabilistically and adding collaborative modes where teams can explore decisions together.

Built With

• css3
• gemini
• google
• google-gemini-api
• html
• javascript
• lovable