Tagline: Production FastAPI backend for Kenyan poultry farmers.
Problem: Broiler farmers lacked data-driven tools to track Feed Conversion Ratio (FCR) and mortality rates, leading to inefficient resource allocation and undetected disease vectors.
Key Engineering Decisions
Raw SQL via asyncpg for sub-30ms FCR aggregation using CTEs and window functions.
Database-level CHECK constraints and triggers enforce zero invalid mortality states.
Numbered SQL migrations for explicit, reviewable schema versioning.
Modular monolith with clean domain boundaries (flock, health, finance).
Case study by Fredrick Nyang'au, Backend Engineer based in Nairobi, Kenya, open to remote roles worldwide. Specialises in FastAPI, PostgreSQL, and M-Pesa Daraja integrations.