- Dive Club
- Posts
- 🤿 the new design deliverable
🤿 the new design deliverable
lessons from one of tech's hottest startups

This week, Zach Leach (Head of Design at Gamma) gives a behind-the-scenes of what it’s like designing one of the hottest AI products in the world today 👀

He shares a ton of techniques and lessons learned from working with LLMs—including some ways to make AI fun again :)
And we also learn what makes design at Gamma a bit unorthodox in all the best ways. Some highlights:
What makes Gamma a team of “AI Pioneers”
How Gamma designers prototype ideas in code
How Zach has iterated on interaction patterns for AI
How to think about AI functionality at the system level
How Zach imagines his design role in the coming years
What Zach has learned about designing AI-powered flows
+ a lot more
🤝 WITH FRAMER
If you’re working at a startup then I have some good news for you…
Framer just released an incredible offer that gives startups free access to their Launch Plan for a year.
That’s $900 to design and ship your site with Framer which is a total no brainer.
I even took advantage of the deal myself 🤷♂️
So definitely don’t miss out on this offer. Click the link to learn more 👇
🎓 KEY TAKEAWAYS
My highlights from how design works a Gamma
1 — Context is now a design deliverable
Zach is actively thinking about how to provide models more context while designing Gamma. This responsibility doesn’t fall solely on engineers.
“giving the models just more context, is always going to turn out better”
Some examples:
Where your cursor is
What’s in your clipboard
What theme the deck is using
What you’ve recently written or deleted
What content is alongside the thing you’re making
etc.
All of this can be fed into LLM-powered features like editing imagery. So a big part of Zach’s prototyping process is piping in different combinations of context and seeing how it affects the output.
2 — Cursor is a design tool
The design team at Gamma has fully embraced AI coding tools like Cursor, Lovable, Bolt, etc…
“We’re a design team who is quite capable with regards to coding, prototyping, etc… we close bugs and ship stuff”
Zach makes an important point that it’s not just about being able to make a 0→1 prototype… AI is also incredible for helping designers understand the code base.
I’m benefiting from this too with Inflight. It’s so empowering to open Cursor and be like “yo this is happening in the product, can you show me where it exists in the codebase and explain how it works?”
“I think to myself… how many bugs can I as a designer just go fix by asking Cursor to do it? It’s kind of mind blowing”
3 — Create loops to test models
Whenever a presentation is generated in Gamma there’s a little faces rating system 😍 😐
Gamma uses this to A/B test different underlying models and they even have an internal leaderboard to see which models are performing the best!
The interesting thing is they can use different models for each individual AI feature. Like maybe one model is best for creating outlines while another one is better for editing images.
Quantitative feedback loops are incredibly important for non-deterministic outputs. So I’m already thinking of ways to add my own 🤔
How much did you enjoy this issue?Never hesitate to reply with feedback too :) |
Meet the Dive partners
I made a list of my favorite products and asked them to come on as sponsors of the newsletter/podcast. They said yes 🥹
The #1 way to support Dive Club is to check them out👇
Framer → How I build my websites
Genway → How I do research
Granola → How I take notes during CRIT
Jitter → How I animate my designs
Lovable → How I build my ideas in code
Mobbin → How I find design inspiration
Paper → How I design like a creative
Raycast → How I stay in flow while I work
Thanks for reading! I'm working hard to bring you the best design resources on the planet 🫶
If you want to go even deeper you can always:
![]() |
|
P.S. if you were forwarded this email you can subscribe here