System Engineering Track Overview
This module is the hands-on track for system engineers and data architects — the Group I stream of the HIC Exchange. Over the week you build the platform end to end: provision the infrastructure, stand up the data warehouse, model its data, and finish by training a real machine-learning pipeline on medical imaging.
What you’ll work through
| Sub-module | Focus | Stack |
|---|---|---|
| Infrastructure | Server and environment set-up, configuration and optimisation, data warehouse set-up, centralised health-facility data ingestion, deployment | KVM, Ansible, Docker, Postgres |
| Data Modeling | Medallion architecture (Bronze / Silver / Gold), modelling DHIS2 aggregates and Maternal Audit data | dbt, Python |
| AI & Machine Learning | Data processing and ML modelling on a real imaging dataset | Python, PyTorch |
A fourth track — Data Ingestion (DHIS2 data elements, DHIS2 Tracker, individual-level data, and web-scraped climate and social-media sources, orchestrated with Prefect) — runs at the event; its tutorial material will be published here when ready.
How the module maps to the event
Each sub-module mirrors a Group I day from the Exchange programme: Environment and Infrastructure Set-up (day 8), Lab 3: Setting Up Medallion Infrastructure (day 10), and Lab 4: AI — Data Processing, ML Modeling (day 11).
Work through the items in order — the labs assume the environment built in the Infrastructure sub-module. Start with Environment and Infrastructure Set-up.