Why I went
all-in on AI.
I spent 4 years building software — Rails backends, React frontends, mobile apps, SaaS platforms. I was good at it. Top Rated Plus on Upwork, $200K+ delivered, a 100% job success score. By every traditional metric, things were going well.
But I kept watching the same pattern repeat: clients would hire engineers to build software that automated something humans were doing manually. A form that replaced a spreadsheet. A dashboard that replaced a report. A scheduler that replaced a phone call. We were automating the symptom, not the cause. The human judgment layer was still there — just one step removed.
When I saw what large language models could actually do — not the demos, but the production deployments — I made a deliberate choice. I rebuilt everything around agentic AI. Multi-agent systems where AI doesn't just process inputs — it reasons, coordinates, decides, and acts autonomously.
I founded Devsphinx to build exactly that for founders and operations leaders who want to eliminate the highest-cost, lowest-leverage work in their business. My full-stack engineering background is what makes the difference. I don't just design agent architectures — I build them to production standard, integrate them with real systems, and deploy them into real operations. The 9,210 hours on Upwork weren't spent on prototypes.
