Project Story: Building a Trust-Powered Digital Committee System

💡 What Inspired Us

In Pakistan, committees (also known as ROSCAs — Rotating Savings and Credit Associations) are one of the most common ways low- and middle-income individuals save money. Families, coworkers, and neighbors pool small monthly amounts, and each month one person receives the full sum.

Despite being widely trusted culturally, the system itself is fragile:

• It depends heavily on a single committee admin
  
• Bias in lucky draws is common
  
• Payments are tracked manually
  
• Defaults are hard to enforce
  
• Finding reliable participants is difficult
  
• There is no formal accountability or transparency
  

Many of us have personally seen or experienced:

• Money disappearing
  
• Organizers favoring friends
  
• Members disappearing after receiving their payout
  
• Entire committees collapsing due to one bad actor
  

This gap between social trust and systemic trust became the core inspiration for our project.

We didn’t want to change the committee system —
we wanted to protect it.

---

🎯 The Problem We Set Out to Solve

Informal savings systems work because of trust, but they fail because there is no infrastructure to enforce that trust.

Mathematically, a ROSCA works perfectly:

[ \text{Monthly Contribution} \times \text{Members} = \text{Payout} ]

But socially, one missed payment can collapse the entire cycle.

We asked:

• What if trust could be measured?
• What if fairness could be verified?
• What if accountability didn’t require confrontation?
• What if committees worked even among strangers?

---

🚀 Our Solution

We built a trust-powered digital committee platform that transforms informal savings into a transparent, accountable, and safe system — without removing its community-driven nature.

Core Ideas:

• Trust Score instead of blind faith
  
• Automated rules instead of manual enforcement
  
• Transparency by design, not promises
  
• AI guidance, not AI decision-making
  

Our platform allows users to:

• Join or create committees digitally
  
• Contribute automatically or track payments
  
• Participate in fair, unbiased draws
  
• Build a public trust reputation over time
  
• Make informed decisions before committing money
  

---

🛠️ How We Built It

Tech Stack

• Frontend: React (modular, dashboard-driven UI)
• Backend: FastAPI (clean APIs, async-friendly)
• Database: MongoDB Atlas (flexible schema for evolving features)
• Payments: Stripe (planned integration)
• AI Assistant: Feature-aware chatbot (Phase 2)

System Architecture

• User profiles with verification layers (CNIC, phone)
• Committee engine with lifecycle states
• Trust score system starting at 5.0, adjusting based on behavior
• AI chatbot that reads data but never controls money

Design Philosophy

• Build structure first, automation second
• Every feature must be explainable to a non-technical user
• No “black box” logic
• AI must assist, not replace judgment

---

🤖 Why We Added AI — and How We Limited It

Instead of using AI for hype, we used it for clarity.

Our chatbot:

• Explains trust scores
• Helps users discover safe committees
• Performs background checks using public platform data
• Guides onboarding step-by-step
• Transparently says “This feature is coming soon” when it doesn’t exist

It cannot:

• Process payments
  
• Modify trust scores
  
• Override rules
  
• Make legal or financial decisions
  

This keeps the system safe, auditable, and fair.

---

⚠️ Challenges We Faced

1. Balancing Trust & Privacy

We needed accountability without exposing sensitive data.
Solution: encrypt sensitive fields, mask public identifiers, and keep AI read-only.

2. Avoiding Over-Engineering

It was tempting to add blockchain, smart contracts, or complex ML early.
We deliberately postponed them to focus on real user value.

3. Designing for Low-Income Users

The system had to be:

• Simple
• Mobile-friendly
• Transparent
• Forgiving but firm
  

Every decision was tested against the question:

“Would this work for someone using it for the first time?”

4. Making AI Honest

Most chatbots overpromise.
Ours had to say:

“This feature isn’t available yet — here’s what you can do now.”

This transparency was a design challenge but became a strength.

---

📚 What We Learned

• Trust is a system problem, not a people problem
  
• Automation should reduce conflict, not increase control
  
• AI is most powerful when it explains, not decides
  
• Fintech products must be humble, especially in underserved communities
  
• Transparency beats features
  

---

🌍 Impact & Vision

Our MVP focuses on Pakistan, but the problem exists globally.
Anywhere people rely on informal savings, our model can apply.

Long-term, we envision:

• A portable trust identity
• Financial inclusion without interest
• Safer community finance
• Reduced defaults through accountability, not fear

---

🏁 Closing Thought

This project wasn’t about building another finance app.
It was about protecting something people already rely on —
and giving it the infrastructure it always deserved.

Trust, made measurable.
Community, made safer.
Savings, made sustainable.

Built With

• async
• atlas
• backend
• cloud)
• created
• database:
• fastapi
• framework:
• framwork)
• frontend
• frontend)
• high-performance
• javascript
• languages:-python-3.10+-(backend)
• mongodb
• motor
• orm/driver:
• react
• via
• vite)