From journalist to app developer using Claude Code

2026-03-16 · Source: How I AI · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, extended

Summary

Dan Roth, Editor-in-Chief at LinkedIn, a non-technical professional, has transitioned into an iOS app developer using AI agents, specifically Anthropic's Claude. He employs a "dueling Claude code" setup featuring two distinct AI personas: "Bob the Builder," who plans and generates code, and "Ray the Reviewer," a security-obsessed senior software engineer who scrutinizes Bob's output for architectural integrity and security. Roth acts as the tie-breaker between Bob and Ray's disagreements. This agentic workflow allows him to develop production-grade iOS apps, such as "Commutely," which helps users track NYC train times. He emphasizes the importance of logging all interactions in markdown files to maintain context and facilitate learning, and he uses a tiered Claude subscription plan ($20 or $100/month) based on his development needs. Roth also leverages Microsoft Copilot for daily work management, using a "What did I drop the ball on?" prompt to track tasks and emails.

Key takeaway

For AI Product Managers or Entrepreneurs looking to accelerate app development without deep technical expertise, consider adopting a multi-agent AI workflow. By assigning distinct personas like a "builder" and a "reviewer" to your AI, you can streamline planning, code generation, and quality assurance. This approach not only speeds up development but also provides a structured learning path, allowing you to focus on product vision and user experience while the AI handles the technical execution.

Key insights

Non-technical users can build production-grade software using a multi-agent AI workflow for planning, coding, and review.

Principles

• Plan thoroughly before building.
• Document everything for context and learning.
• Assume best intentions when managing AI.

Method

Utilize distinct AI agents for specialized tasks like building and reviewing. Log all interactions and decisions in markdown files. Manually transfer outputs between agents to foster learning and critical oversight.

In practice

• Create AI personas for specific development roles.
• Use markdown files to track project ideas and AI interactions.
• Implement a 'picky customer' role for app refinement.

Topics

• AI Agentic Workflow
• Claude Code
• iOS App Development
• AI-Powered Software Engineering
• Non-Technical AI Adoption

Best for: Software Engineer, AI Product Manager, Entrepreneur

Related on AIssential

• Claude Can Make Hundreds of AI Agents Now! 🤯 — Claude Code's workflow automates complex multi-agent development, planning, and rigorous QA testing.
• How I code with agents, without being 'technical' — AI agents enable non-technical users to build complex software by abstracting traditional coding into a guided, iterative process.
• I Built My Friend a Free Personalised Fitness Plan Online — Without Writing a Single Line of Code — AI can rapidly generate and deploy functional web applications from detailed prompts.
• How a non-technical project manager built and shipped a stress management app with Claude Code in six weeks — Non-technical users can rapidly build and ship complex applications using AI code assistants like Claude Code.
• Donchitos / Claude-Code-Game-Studios — Structure AI game development with a multi-agent system for organized, collaborative, and quality-controlled workflows.
• Eight years of wanting, three months of building with AI — AI excels at tedious coding tasks but struggles with high-level architectural design and subjective problem-solving.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by How I AI.