Neura-Q

Inspiration - In a world dominated by AI, where I'm a heavy user myself, I noticed a trend — we often just accept AI recommendations and go with them blindly. But decision-making isn't just about choosing; it's about thinking through the dilemma. That sparked the idea:

What if we could build an AI that doesn’t just give you an answer — but helps you think better? That’s how Neura-Q was born — a quantum-thinking AI that helps you explore every perspective, understand trade-offs, and think like high-stakes decision-makers.

What it does - Neura-Q is an AI-powered decision-thinking tool that lets users map out complex dilemmas using different structured reasoning frameworks. It offers three intelligent agents:

Aeon – For career dilemmas. Aeon uses Tree of Thought (ToT) reasoning to break down long-term impacts and strategic options.

Nova – For life decisions. Nova uses Chain of Thought (CoT) reasoning and adapts to your chosen persona to think in a specific mindset.

Tess – For business/startup dilemmas. Tess uses Graph of Thought (GoT) to map interconnected risks, outcomes, and lets you define your own Success Matrix to tailor answers to your personal goals.

Together, they help you not just choose — but understand your decision landscape.

How we built it - We started building the app using bolt.new for rapid prototyping and design flow. For the backend, we used Supabase Edge Functions to handle different AI modes and securely store user sessions.

Each agent is powered by structured prompting and Gemini Pro for reasoning. On the frontend, we used React + Tailwind CSS with D3.js for rendering beautiful, interactive visualizations (trees and graphs) that make your thinking tangible. We ensured smooth user experience using shadcn/ui components and Framer Motion for animated transitions.

Challenges we ran into -

Building clean and meaningful visualizations for complex thoughts (like ToT and GoT) was a huge challenge. Balancing clarity and aesthetics was tough.

Prompt engineering took a lot of effort to make sure each AI agent behaved differently and logically.

Handling dynamic user input and passing it to Supabase Edge Functions reliably across modes required deep backend coordination.

Auth integration and session persistence had some hiccups, especially across Vercel and Supabase environments.

Accomplishments that we're proud of - We’ve built a thinking engine — not just a chatbot.

Neura-Q helps people structure their thoughts, map out options, and gain deeper insight before deciding. We’re proud of the visual clarity it brings to abstract decisions and the flexibility to think through your goals using powerful AI modes.

What we learned - We learned how to design a pipeline that goes beyond "ask and answer" — to build logic-based, mode-specific AI tools.

We deepened our understanding of frontend-backend interaction, prompt design, user experience, and integrating Supabase Edge Functions. And most importantly — we learned how complex it is to make decision-making feel simple.

What's next for Neura-Q - We’ll be refining the product further, including:

• Introducing a stronger visual recommendation path in the graphs

  • Adding detailed note explanations per node
  • Improving AI accuracy and persona consistency
  • Possibly moving to our own backend for more control
  • Exploring voice input and mobile-first UX

Neura-Q is just getting started — the future of decision intelligence is visual, structured, and deeply personal.

Built With

• bolt.new
• d3.js
• framer-motion
• google-gemiini-api
• shadcn/ui
• supabase