When people talk about AI quality, they often focus on models.

I usually look at inputs first.

Poor inputs create noisy outputs.

Clear inputs often improve results more than adding complexity.

That applies whether writing prompts, structuring data, or designing automation.

A simple example

Instead of:

summarize(notes)

I might provide stronger context:

summarize({
  notes,
  audience: "technical",
  length: "short"
})

Same model.

Better constraints.

Better result.

I usually think about AI systems in terms of:

• input clarity
• useful constraints
• predictable outputs