Most Product Designers are still living in frames.
I'm building directly in the product.
Not because I'm a developer.
But because the tools changed and I changed with them.
Here's my actual workflow right now:
→ Clone the existing product's repo
→ Create a branch (safe to experiment, nothing breaks)
→ Run a UX audit on what's already there
→ Identify exactly where an agentic AI layer adds real value
→ Use Claude Code to ship the enhancement directly
I'm not redesigning things in Figma first.
I'm going straight into the codebase and making the product smarter.
This is what "agentic AI layer" actually means not a new product built from scratch. Your existing product, with AI that thinks and acts on behalf of your user.
The result for businesses?
Faster time-to-value. No need to rebuild.
Your current product becomes dramatically more intelligent without throwing away what already works.
The biggest shift for product teams right now isn't hiring AI developers.
It's working with someone who understands both
the user experience AND how to inject AI that serves it.
Frames were never the final product anyway.
If you have an existing product and want to explore
where an agentic AI layer could make the biggest impact
DM me. Let's talk.
What sells it for me is the refusal telling the agent exactly where the token should go. A bare "no hex" error usually just sends it looping. Does guard.mjs run only as an agent hook, or in pre-commit too so it catches humans as well?
I'd pick Violet Desk. Blue is what every SaaS uses, violet stands out and still says AI, and the orange sticky notes pop more against it. Does the hero chip row under the headline work on mobile?