A new article making rounds on Hacker News offers a refreshingly practical perspective on using AI for design work—from the viewpoint of someone who explicitly does not consider themselves a designer. Titled 'How I Design with AI as an Engineer Who Is Not a Designer and Hates Slop,' the piece cuts through the glossy demo videos to address the messy reality developers face when trying to incorporate generative tools into their workflow. The author tackles a problem many in the engineering community have grappled with: AI design tools often produce output that looks impressive at first glance but falls apart under closer inspection. 'Slop'—the pejorative term for low-quality, generic AI-generated content—has become a significant concern as these tools proliferate.

Building a Workflow That Respects Your Constraints

For developers who find themselves needing to handle UI or visual work without formal design training, the challenge isn't just about making things look good—it's about building repeatable processes that don't require starting from scratch every time. The article suggests establishing patterns and constraints that guide AI outputs toward useful directions rather than letting models generate whatever looks visually coherent but contextually wrong. The approach described appears to emphasize treating AI as a tool within an existing workflow rather than a magic button that produces final deliverables. This aligns with what many experienced practitioners have learned: generative systems work best when human judgment defines the boundaries and makes final decisions.

The Honest Assessment We Need More Of

What's notable about this piece is its self-aware framing. By leading with 'I am not a designer,' the author sidesteps imposter syndrome and focuses on what's actually useful—sharing techniques that work regardless of background. This kind of practitioner-to-practitioner knowledge transfer tends to be more valuable than polished marketing content or academic treatments of AI capabilities. The discussion on Hacker News had minimal engagement at time of reporting, suggesting the topic may be finding its audience more through search and direct sharing than viral spread. The full article is available at ref.tools.

Key Takeaways

  • Non-designer developers can leverage AI tools effectively by establishing clear constraints upfront
  • Avoiding 'slop' requires active guidance rather than passive prompting
  • Treating AI as a workflow component beats treating it as an autonomous solution
  • Community-driven insights from practitioners often outperform official documentation

The Bottom Line

Being practical about AI capabilities means acknowledging both what these tools can do and where they still require human judgment to avoid producing polished-looking but ultimately useless output. For developers building products without dedicated design resources, approaches like those outlined in this piece offer a realistic path forward—less hype, more craft.