Built by a test engineer who got tired of AI shipping untested.

SPERO is led by Douglas Bonafe, a software and test engineer with 15+ years in QA automation and a Master's in Computer Engineering from UNIFEI.

For most of that time the job was simple to explain: find what breaks before the customer does. When teams started shipping LLM features, that discipline mostly disappeared. Prompts changed without tests. Models were swapped without comparison. Quality was judged by reading a few outputs.

SPERO exists to bring engineering discipline back to AI features. Datasets, evals, regression gates, the boring stuff that makes things work.