Lior A.

Most clients come to me with a symptom, not a solution. The manual work that's eating 15 hours a week. The data living in 4 different places. The process that breaks every time someone new joins. My job is to figure out what's actually worth building before we build anything. I don't need a full spec. Tell me what's breaking or what's taking too long, and I'll tell you whether it's worth automating, what it would take, and what to build first. That instinct comes from running a business, not just building for one. I founded and scaled a disaster restoration company to $1M+ in annual revenue, managing 90+ projects and crews of 5 to 15. When your livelihood depends on operations running smoothly, you get very good at spotting where systems are failing and what actually fixes them. Recent projects: an automated tracking system that cut 10 to 15 hours of weekly manual data entry for a construction supply chain, a subcontractor operations platform for a 900-unit property management company, and a custom underwriting system integrating 4 credit bureau APIs for a $4M+ lending portfolio. In each case, the first conversation wasn't about code. It was about what was actually costing time and money. My current stack includes Claude API, MCP servers, n8n, GoHighLevel, Python, Firebase, and React. I hold Anthropic certifications in Claude Code in Action and MCP Architecture, both issued April 2026. If you have a clear spec and just need execution, I'm probably not the right fit. If you know something is broken but aren't sure what to build, that's exactly where I start.