The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
AI Agent Designer work starts with the channels, not the prompt.
For MeetClaw, I designed the setup around where the assistant needs to live, which model provider fits, and where state should stay. The work then moves through installation, configuration, and hardening on the client's own machine.
I build with approvals, read-back, and real integrations in the loop, so the design survives contact with the workflow instead of stopping at a diagram.
If your agent needs to work across several channels, message me with the machine and the channels.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
AI draft → my direction 🌿
I gave Figma Agent the content, structure, and references to create a quick starting point, then redesigned it in my own visual direction.
AI for speed. Creative decisions by me.
Before or after? 👀
Beautiful
A workflow can finish successfully and still be wrong.
That is the reliability problem with production AI agents. A green execution only tells you the workflow ran; it does not tell you whether the model had the right context, chose the right tool, or returned a useful answer.
A practical n8n reliability loop:
• constrain inputs, tools, schemas and routing
• tag executions and inspect full agent logs
• run evals whenever prompts, tools or models change
• add real production failures to the test set
• monitor execution, quality, efficiency and safety separately
The best metric is not the one you can collect. It is the one that changes a decision.
Which failure would your current monitoring miss?
For this Inplac project, the starting point was pretty limited: the client had older product mockups that showed the packaging, but not much beyond that. No real marketing treatment, no lifestyle context, no visual story around the product.
So instead of simply polishing what already existed, we used AI to completely rethink how those products could be presented.
We created new product visuals, placed them into more realistic and human contexts, and then pushed those scenes further through animation, turning static mockups into short pieces that felt much closer to actual campaign content.
The biggest challenge was keeping the product recognizable and consistent while changing everything around it: environment, lighting, composition, scale and movement.
That’s the part of AI work I find most interesting. It’s not about generating something “cool”. It’s about starting with very little and building a visual that actually helps the product communicate better.
