# How I used AI to build a slop-free web page

## Executive summary

The speaker details a process for achieving high-quality web design for the Multiplayer AI Manifesto page, emphasizing that while AI was used, the core content and creative direction were human-driven. The process involved writing all copy manually, specifying a desired 'vibe,' and utilizing a 'random-string trick' in prompts to generate design variety. The site was built by launching multiple AI agents (including Fable 5.1) on the Superconductor platform, comparing their screenshots, and then engaging in extensive, iterative, multi-agent chat sessions to refine the final product.

## Key takeaways

- Content Creation: All copy was written by hand, and the desired visual 'vibe' was specified in the source document, avoiding AI writing for the text.
- Design Variety Trick: To achieve design diversity, the speaker used a 'random-string trick' in the prompt, instructing the AI to generate a long alphanumeric string and apply it to the design without mentioning the string itself.
- Multi-Agent Testing: The process involved sending a Notion link to Superconductor to build the site, requesting a sequence of screenshots using Playwright. The speaker launched 14 different AI implementations (e.g., Fable 5.1, CodeXes, open-source models) to compare results.
- Iterative Refinement: The final site was refined through a long, multiplayer AI chat session (333 messages), where the speaker and co-founder provided visual feedback (via screenshots) to the chosen agent (Fable 5.1) for continuous improvement.

## Technical details

- AI Agents & Platforms: The site was built using Superconductor, which allowed for launching and comparing multiple AI implementations (e.g., Fable 5.1, CodeXes, Chinese model).
- Testing & Automation: The agents generated screenshots using Playwright to allow for visual comparison and selection of the best design.
- Prompt Engineering: A key technique was using a 'random-string trick' in the prompt to force design diversity, rather than relying on generic 'build me a site' prompts.

## Practical implications

- For build engineers, the process highlights that AI can be used as a rapid prototyping and design iteration engine, but human oversight (writing copy, defining the 'vibe,' and providing iterative feedback) remains critical for quality.
- The use of multiple agents and automated screenshot capture (Playwright) provides a scalable way to test design hypotheses quickly.
- The 'random-string trick' demonstrates a sophisticated prompt engineering technique to break AI from generating predictable, generic designs.

## Topics

AI, Web Development, Prompt Engineering, Design Systems, Multi-Agent Systems, Superconductor, Fable 5.1, Playwright, Notion

Source: https://www.youtube.com/watch?v=eyXONucnUCA