Opinions held loosely, convictions held firmly.
We change our minds in the face of evidence. The things we believe in — clarity, maintainability, respect for the people who will run this system at 2 a.m. — those, we will defend.
Novexa is an independent software consultancy, incorporated in the United States, serving clients across North America and Europe. We are deliberately small, intentionally narrow, and we take a long view of every engagement.
Novexa began with a simple conviction: that most software problems are not technology problems, they are judgment problems — and judgment is what a good consultancy is actually selling.
We founded the practice to work alongside engineering teams on the parts of their stack that carry the most weight. Cloud foundations that have to stay up. AI features that have to be honest. Technical decisions whose consequences will still be felt in five years.
Our clients are not buying hours. They are buying the benefit of having seen this situation before, of having worked through the specific failure modes, of knowing where the bodies are buried. They are buying calibration.
We stay small because small is how we stay good. We choose every engagement carefully, we say no more often than we say yes, and we are never, ever too busy to do the thinking part of the work.
We change our minds in the face of evidence. The things we believe in — clarity, maintainability, respect for the people who will run this system at 2 a.m. — those, we will defend.
We reach for the proven tool before the fashionable one. Novelty is a cost. It earns its place when it genuinely outperforms the incumbent, not when it is merely interesting.
If a plan is wrong, we say so — early, directly, and with our reasoning. Consulting that agrees with everything is a form of malpractice.
Our goal is never to become indispensable. It is the opposite: to hand over, to document, to teach, so that the capability stays with your team when we go.
Leo has spent the last fifteen years working at the boundary between platform engineering and applied AI — first as a staff engineer at a series of B2B SaaS companies, later as principal consultant for funded startups and enterprises navigating cloud migrations, regulated AI deployments, and the move from monolith to durable platform.
Earlier in his career he led infrastructure at a mid-stage fintech through SOC 2 and ISO 27001 readiness, designed the first generation of a multi-tenant data platform now serving several hundred enterprise customers, and shipped retrieval systems in production long before they were called RAG. He founded Novexa to do that work on his own terms, in small, carefully chosen engagements.