MindCouncil-AI

Your Personal Multi-Perspective Decision Assistant

Comment

MindCouncil AI is a personal multi-perspective decision assistant designed to help users make complex choices with confidence. The inspiration came from realizing that many people face difficult dilemmas in both personal and professional life, but rarely have access to structured, unbiased advice. Traditional AI assistants often give generic answers, so I wanted to simulate a council of diverse personas, each offering a unique perspective.

Inspiration

I was inspired by the idea of combining AI reasoning with human-like council deliberation. Imagine having a rational analyst, an emotional self, a risk manager, a future self, and a devil’s advocate all debating your decision simultaneously — that’s what MindCouncil AI aims to achieve.

What I Learned

Building modular AI services for each “persona” and integrating them in a single backend.

Using Flask + JavaScript frontend to create a responsive dashboard.

Implementing decision history, confidence scoring, and scenario simulation.

Handling asynchronous API calls and state management in a multi-step workflow.

How I Built It

Backend: Flask-based microservices, each handling one step of the decision-making process. Integrated OpenRouter AI (Gemini 3) for NLP reasoning.

Frontend: HTML/CSS/JS dashboard that guides users through 12 steps — from dilemma input to action plan generation.

Persona Simulation: Prompt engineering and a small council engine to mimic multiple reasoning perspectives.

State & Memory: Session-based storage for decision tracking and history.

Challenges Faced

Managing multiple asynchronous API calls without freezing the frontend.

Designing clear, intuitive UI/UX for 12-step complex workflows.

Handling errors, missing data, and ambiguous user input gracefully.

Integrating AI reasoning while keeping outputs explainable and actionable.

Built With

  • css
  • css-(frontend)-frameworks-&-libraries:-flask
  • flask-cors
  • github-pages-/-netlify-(frontend)-environment-&-config:-python-dotenv-for-environment-variables
  • html
  • httpx
  • javascript
  • jinja2-apis-&-ai-services:-openrouter-ai-(gemini-3-flash-preview)-frontend-tools:-vanilla-js
  • languages:-python-(backend)
  • pydantic
  • session-&-memory-management-for-state-persistence-version-control:-git
  • tailwind-(if-using)-hosting-/-deployment:-render-(backend)
Share this project:

Updates