Tag
All Nuvika Blog posts tagged “Engineering”.
Engineering
A runtime env-var fix that worked every time and broke every deploy. Why imperative patches to a declaratively-owned Cloud Run service have a half-life of exactly one deployment.
Engineering
A long-lived git worktree was quietly running on a months-old copy of CLAUDE.md. Why copies don't subscribe to updates, and the discipline that prevents silent context drift.
Engineering
A technical founder's honest account of shipping a complex multi-provider feature: what we designed, what we got wrong in code review, and what we'd do differently.
AI
How we used Claude Code to implement, review, and verify a complete feature end-to-end — TDD, subagent-driven development, two-stage code review. What worked, what surprised us.
Cloud
The rule engine architecture behind Fintropy's cost optimisation scans — deterministic Tier 1 rules, AI-assisted Tier 2 rules, and a registry that auto-discovers both.
Engineering
The error handling design behind Fintropy's auto-claim filing — why a filing failure must never break a breach resolution, and how we guarantee it.
Fastapi
Why Fintropy chose FastAPI over Django for its 84-endpoint backend, and the honest cost of that decision a year and a half in.
Postgresql
Why Fintropy chose PostgreSQL Row-Level Security over separate schemas or databases for multi-tenancy, and what it actually costs.
Cloud-Architecture
How Fintropy switched from Kubernetes to Cloud Run, what we gained, and the cold-start SIGTERM problem nobody warned us about.