Government Digital Observatory : Phase 2

An earlier phase had already produced a working version of the observatory. What was still open was turning it into a system a national institution could actually operate day to day, in its own infrastructure, rather than software that merely ran. That gap defined the phase 2 mandate.

The brief split into two halves that had to ship together: the application a Django/DRF API and a Next.js frontend, with PostgreSQL/PostGIS holding the observatory's geospatial data and the infrastructure it would run on: Docker images for every service, a production Kubernetes cluster, and a reverse proxy terminating HTTPS on-site. Because deployment targeted the client's own national datacenter rather than a managed cloud, the infrastructure half wasn't optional scaffolding it was as much the deliverable as the application.

Led the phase as lead developer: designed and built the backend API and the Next.js frontend, containerised every service, and deployed the cluster in production.

That included on-site DevOps missions at the client's national datacenter in West Africa standing up the reverse proxy, HTTPS/SSL and persistent storage in place, rather than through a remote pipeline. Alongside the build, led the technical team, collected the observatory's data, trained the end users, and handed over documentation and support.