Jonathan Gur-Zeev | Weavy.ai | How to Build a “Design Machine” for Organizations in the AI Era - Product Builder Podcast episode with Eyal David
Updated: 00:53:30

Jonathan Gur-Zeev | Weavy.ai | How to Build a “Design Machine” for Organizations in the AI Era

Host: · Guest: Jonathan Gur-Zeev · About the host

Jonathan Gur-Zeev, co-founder and CPO of Weavy.ai and a former product group lead at Fiverr, explains why AI has barely entered professional image and video production: results are inconsistent, there's no control, and the tools are scattered. Weavy's answer is a "design machine": instead of designing every frame, the designer builds a machine that combines AI models with editing tools and produces endless outputs according to his own design decisions.

In this episode

  • AI won't enter professional production on speed alone. Weavy's goal isn't working twice as fast but building a "design machine" that produces a thousand times more output without giving up control and professional integrity.
  • What separates a professional from a hobbyist is that there's a client at the end. So professionals need consistency, control and editability, not just a result that looks good.
  • At the zero-to-one stage it's better to build something real, even if shaky, and let selected users work on it than to show mockups. Watching usage recordings teaches more than what users say.
  • Don't rush to launch a product for designers. When you move a professional audience's cheese, a bad first impression is hard to fix, so launch only when you're not embarrassed by the product.
  • The way into organizations is bottom-up: designers who love the tool, managers who save on licenses for many tools, and a view for the machines' users on marketing teams.
  • A good product manager is measured by agency and getting things done, not by knowing methodologies. Knowing you don't know everything and wanting to learn matters more than Jira and Scrum.

Listen to this episode

Listen on SpotifyListen on Apple PodcastsWatch on YouTube

Episode Description

Jonathan Gur-Zeev, one of the four founders of Weavy.ai, describes how the realization that design and video professionals use AI only for inspiration and moodboards led to a platform that combines any AI model with traditional editing tools. He explains the "design machine" idea, the challenge of AI FOMO and AI literacy, and the decision to build a flexible platform rather than solve a single use case. He then describes how the team skipped mockups and worked directly with real users, how a LinkedIn post brought NVIDIA to them, why first impressions matter most, and why in his view intuition beats methodology and agency is the most important trait of a product manager.

Chapters

  1. 0:28 · What Weavy is and why AI isn't in production yet
  2. 5:44 · From Fiverr to Porto: how the founding team came together
  3. 7:48 · The problem: designers use Midjourney only for inspiration
  4. 9:45 · Consolidating tools, and a "design machine" instead of manual design
  5. 15:01 · AI FOMO, a "Zapier of AI" and AI literacy
  6. 18:22 · One use case, or a platform for every use case
  7. 21:36 · Skipping mockups: design partners and real users
  8. 25:27 · The LinkedIn post that brought in NVIDIA
  9. 36:11 · Intuition vs. methodology
  10. 42:56 · First impressions and entering organizations bottom-up
  11. 47:09 · The product skills that matter: agency and getting it done
Read the full transcript
Full transcript with speakers and timestamps

Selected quotes

“Instead of designing or instead of editing, build a machine that designs and edits for you.”
— Jonathan Gur-Zeev
“If you have good intuition, you don't need a methodology.”
— Jonathan Gur-Zeev
“It'll launch when I don't feel embarrassed to launch it.”
— Jonathan Gur-Zeev

Questions & answers

What is Weavy.ai?

Weavy is an image and video creation platform for professionals in marketing, gaming and branding. It combines existing and future AI models with traditional editing tools, letting you build a "design machine" that produces many outputs according to the designer's own design decisions.

Why hasn't AI entered the production work of designers and video editors yet?

According to Jonathan Gur-Zeev, AI results are non-deterministic and there isn't enough control to get consistent outputs a client will approve. In addition, the tools are scattered across dozens of products with different interfaces and terms, so most professionals use AI only for inspiration and moodboards.

What is a "design machine"?

Instead of designing a single frame or shot, the designer builds a process that combines AI models and editing tools and defines composition, color, details and depth of field. The machine runs again and again and produces unique outputs that always stay true to the guidelines the designer chose.

How did Weavy do discovery without mockups?

The team built a real early version and brought in people they knew one by one, with personal onboarding, interviews in front of the product and watching usage recordings. Today they work with design partners and freelancers to validate assumptions faster.

How did Weavy reach NVIDIA?

Even though they asked users not to post, one user shared a screenshot on LinkedIn. Following the post, an NVIDIA studio in the US with about 200 people joined the waitlist, and Weavy ran demos for groups of five to ten people at a time.

People & resources mentioned

People

Tools

generative AI design and video tools for designers startup discovery go-to-market onboarding agency

Related Episodes