The original title is "Building a company in 2026: What I'm learning along the way"

Short summary

A second-time founder and former CTO shares lessons from building CatechLabs in 2026, where AI has made shipping software fast but building a viable business remains hard. The key shift is moving from technical features to workflow-level value, illustrated when a customer asked why they would pay for the platform instead of just giving their team access to Claude. The post emphasizes customer discovery, real buying signals over polite praise, and learning sales and positioning alongside engineering.

• •AI makes building software fast but does not solve for product-market fit or trust
• •Customer conversations should focus on workflow pain and buying signals, not product demos
• •Positioning shifted from model capabilities to structured workflows, audit trails, and team-level repeatability

Generated with AI, which can make mistakes.