How I Design Systems
These diagrams show how the systems I build are actually structured. From field data collection in Nairobi's settlements to payment processing and World Bank reporting — this is the architecture behind the numbers.
System Overview
The bird's-eye view. Data flows from the PWA frontend through API routes into a 76-table PostgreSQL database with PostGIS, with separate pathways for payment processing, government reporting, and field data ingestion from ODK Central.
Field Data Collection Pipeline
How building footprints, survey responses, and 360-degree imagery move from field devices through my validation pipeline and into the central database. Invalid records get flagged by the AI Guardian system and routed back for correction before they reach stakeholder dashboards.
Payment Processing Flow
The stipend and payment pipeline. Field agents submit work reports, which are validated against task completion data. Payments route through either M-Pesa (for KES stipends) or the Lightning Network (for Bitcoin-based programmes), with all transactions recorded to an auditable ledger.
Technology Stack
The tools and frameworks I use in production every day, grouped by layer. Nothing here is aspirational — each one is backed by real project work.
Frontend
The client-facing layer, built for performance and offline resilience.
Backend
Server-side logic handling API routing, authentication, and business rules.
Data
Geospatial-aware storage and survey data management.
Infrastructure
Deployment, containerization, and continuous delivery pipeline.
Security
Access control, encryption, and data protection compliance.
Tools
GIS editors, survey platforms, and developer tooling used across projects.