He Uses 7 Claude Code Agents to Build Apps with 0 Employees

Gabor’s core claim is that better outcomes come from role specialization (system analyst, CTO, designer, test architect, maintainability reviewer) and structured handoffs, rather than asking one model to “build the whole app” in one shot.

He uses the system analyst to force clarifying questions, break requirements into structured documentation, and generate Jira tickets; this reduces ambiguity, prevents rework, and gives coding agents unambiguous tasks.

Gabor argues many “vibe coding” failures are maintainability failures; he adds an explicit maintainability/spaghetti agent to check naming conventions, circular references, and code structure—especially important for non-engineers who can’t easily detect architectural rot.

He notes that when he didn’t break design work into smaller tickets, the design underused parts of the palette from the style guide—his hypothesis is that long context windows still lead to prioritization/compression that drops constraints.

By storing decisions in Confluence and turning them into Jira work items (with dependencies and sprint tags), he makes the build reproducible and easier to iterate, similar to how teams maintain velocity over time.

He emphasizes attaching screenshots or deep Figma links per ticket; otherwise coding agents tend to produce “default AI-looking” UI rather than implementing the intended design system.

He repeatedly constrains key handling (Firebase Secret Manager only), limits user usage (20,000 words then 24-hour cooldown), and warns to scrutinize Claude Code permissions—especially requests outside the project folder (e.g., password stores).