From One Act of Kindness to a Global Movement
A stranger's simple act of helping someone find a home sparked a journey that would become DMAFB. Founded on the principle of "I help you, you help me. Everybody happy" - we're building a digital wellbeing platform that transforms micro-acts of kindness into tools for connection, cultural exchange, and mental wellbeing imp
rovement.
Our Mission: Prevent burnout, improve workplace morale, and make kindness a natural part of daily work life through science-driven, human-first interventions.
What we're building
We're developing an MVP platform that combines:
* Predictive analytics for early detection of workplace wellbeing issues
* Real-time monitoring and personalised intervention recommendations
* Time-series tracking of individual and team wellness metrics
* Pattern recognition to identify hidden risks before they become crises
* A lightweight, intuitive interface that seamlessly integrates into daily workflows
This isn't just another wellness platform - it's a proactive system that shifts organisations from reactive surveys to meaningful, data-driven support.
What we need
We're seeking a volunteer Full Stack Developer with experience in:
Recommended Tech Stack
Backend
Primary Framework Options:
* Python with Django/FastAPI - Ideal choice given the ML/AI requirements (predictive modeling, anomaly detection with auto-
encoders)Node.
* js with NestJS - Good for real-time data processing and high concurrencyJava/
* Kotlin with Spring Boot - Enterprise-grade with strong security
Recommendation: Python with FastAPI
o Excellent for ML integration (scikit-learn, TensorFlow, PyTorch)
o Fast performance with async capabilities
o Native support for data science libraries
o Easy API documentation with OpenAPI/Swagger
Database Architecture
Primary Database:
o PostgreSQL - ACID compliance, excellent for complex queries and
analyticsSuppo
o rts time-series data for tracking trends over weeks/months/quarters
Additional Data Stores:
+ Redis - Caching layer for real-time scoring and dashboard pe
rformanceTimes
+ caleDB (PostgreSQL extension) - Optimized for time-series wellbeing dataElast
+ icsearch - Fast searching through historical patterns and anomaly detection
Machine Learning & Analytics
# Python ML Stack:
# scikit-learn for predictive modeling
# TensorFlow/PyTorch for auto-encoder neural networks (anomaly detection)
# pandas/NumPy for data analysis
# SciPy for statistical analysis
Real-Time Processing
# Apache Kafka or RabbitMQ - Event streaming for survey responses and intervention triggers
# Celery - Asynchronous task queue for scheduled interventions and alerts
Frontend
# React or Vue.js - Interactive dashboards
# D3.js or Chart.js - Data visualizations
# Material-UI or Tailwind CSS - Component library
Security & Privacy (Critical for HR Data)
# OAuth 2.0 / OpenID Connect - Authentication
# Role-Based Access Control (RBAC) - Manager/employee/HR permissions
# End-to-end encryption for sensitive employee data
# Audit logging for compliance (GDPR, HIPAA if applicable)
Cloud Infrastructure
Recommended: AWS or Azure
# Compute: ECS/EKS (AWS) or AKS (Azure) for containerized
servicesStora
# ge: S3/Azure Blob for documents and resourcesCDN:
# CloudFront/Azure CDN for fast resource deliveryMonit
# oring: CloudWatch/Azure Monitor + DataDog or New Relic
Architecture Pattern
Microservices Architecture:
* Survey Service - Data c
ollectionAnaly
* tics Service - Scoring and pattern recognitionInter
* vention Service - Triggered actionsNotif
* ication Service - Email/in-app alertsDashb
* oard Service - API for frontendML Se
* rvice - Predictive modelling
Key Technical Considerations
For the Item-Level Anomaly Detection:
o Auto-encoder neural networks to detect unusual patterns
o Real-time scoring with <100ms response time
o Batch processing for trend analysis
For Predictive Modelling:
o Time-series forecasting models (LSTM, Prophet)
o 1-4 week prediction windows
o Continuous model retraining with new data
For Privacy:
o Data anonymisation at the database level
o Aggregation services that prevent de-anonymisation
o Separate data stores for identifiable vs. anonymous data
What you'll contribute to
o Core platform architecture for our MVP launch
o Integration points for predictive models and analytics
o User-facing features that make wellbeing interventions effortless
o A scalable foundation that can grow with our community
What you'll gain
✨ Purpose-driven work - Help prevent burnout and improve lives globally
✨ Ground-floor opportunity - Shape the technical foundation of a growing movement
✨ Collaborative team - Work alongside ML specialists and wellness experts
✨ Portfolio value - Real-world experience building AI-integrated healthcare tech
✨ Potential equity - Contributors will be considered for equity opportunities