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
"Software engineers who don’t know how to use AI coding agents will fall behind."
No.
AI coding agents are not the hard part of building software. They’re a simple tool. You can learn to use them in a few days.
The hard part is knowing what to ask for.
Knowing whether the answer is any good.
Knowing when the code is brittle, overcomplicated, insecure, or just plain wrong.
It’s the judgement required to use them well.
That’s software engineering. And that takes decades to learn properly.
The people at risk aren’t engineers who haven’t mastered using coding agents yet.
It’s people who only know how to prompt one.
The judgment point is what gets missed in every "AI will replace X" take. Tools change the workflow, not the expertise needed to evaluate the output. Same applies in design.
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
Related posts
Designed an AI Coding Agent Heading, How is it?
Open for feedback🙌
Great job!
Meet Your AI Agent, my entry for the #GoogleStitchChallenge
It's an app for designing what your AI agent or bot looks like. The theme was "build interfaces that feel alive," so I made something where anyone can pick a body, face, head, and aura, name it, give it a personality, and watch their bot come to life. All with zero code.
How I built it with Stitch:
I started from one prompt describing the idea, then built it up conversationally. I added native motion (the bot floats, blinks, and powers up), an animated shader nebula background, and hover states. Whenever something was off, I didn't touch any code. I just pointed at the element on screen and edited it directly, or described the change in plain words.
What I loved: when it's done, Stitch lets you publish in lots of places like Google AI Studio, Figma, MCP, Bolt, Lovable, or Netlify. I chose Netlify because it's so easy to deploy on the web. One click and you get a live link to share.
Stitch features showcased: native motion, animation and shaders, in-place AI edits, and publishing to Netlify.
As someone who can't really code, Stitch never made me feel stuck. It turned an idea in my head into a real, interactive product in a single sitting.
Try it and design your own agent: https://meetmyaiagent.netlify.app/
Walkthrough video:👇
Built with @stitchbygoogle https://stitch.withgoogle.com/projects/6351891533339790820
Fun idea. Giving people control over the personality and appearance of their AI agent makes it feel much more personal than a typical chatbot experience. Curious which customization option people seem to spend the most time on?
I explored a different direction for the challenge:
https://on.contra.com/pnfJBQ
on.contra.com
Introducing Voya: Revolutionize Your Travel Planning Experience
Connect with next-gen talent and tools to get work underway. Hire more independents. Start more projects. Get more creative.
Aura - The World's First Emotion-Adaptive Interface System
Designed with Google Stitch.
As a developer, I've used plenty of AI tools that can generate screens, but I was curious about something different with Stitch:
Could I use it to create an experience-first interface that felt alive rather than just another website?
That question became AURA — an experimental emotion-adaptive interface that transforms its visual environment based on how users want to feel.
Instead of approaching the challenge as a traditional landing page, I wanted to explore a more immersive direction. What if an interface could behave like a living environment? What if changing an emotional state could alter the atmosphere, motion, lighting, color palette, and overall experience?
To test that idea, I designed AURA as a digital experience where users can explore emotional worlds such as Joy, Peace, Wonder, Creativity, Ambition, and Reflection.
How Stitch fit into the workflow
I didn't start with prompts, I started with ideas.
The first step was sketching rough wireframes and visual concepts to establish the structure and flow of the experience. I then created a custom design.md document containing the creative direction, emotional worlds, visual principles, motion goals, typography, color systems, and overall artistic vision for the project.
After importing those references into Stitch, the process became incredibly fluid. The streaming generation experience was one of the most impressive parts of the workflow. Instead of waiting for a final output, I could literally watch the interface take shape directly on the canvas and react to ideas in real time.
Once the foundation was generated, I spent most of my time using in-place AI edits.
I'd select sections, experiment with different visual treatments, adjust atmosphere, refine layouts, introduce new interactions, and push the experience closer to the vision I had in mind without having to restart from scratch.
That iterative loop felt much closer to actual design work than traditional prompting.
Motion was another area where Stitch stood out.
Since the challenge theme was "Build interfaces that feel alive," I focused heavily on creating smooth transitions, hover interactions, atmospheric movement, environmental effects, and immersive visual storytelling.
The result feels less like a conventional website and more like a digital art installation.
New Stitch features I leaned on
📄 Imported custom design.md creative direction document
📄 Imported rough wireframe concepts and visual references
🌊 Streaming generations directly to the canvas
⏩ In-place AI edits using prompts and point-and-click interactions
🎥 Motion-rich interactions, hover states, and immersive transitions
🎨 Rapid visual experimentation and iteration
🚀 Export-ready workflow for deployment
Feedback on Stitch
Most AI workflows involve repeatedly prompting, waiting, evaluating, and starting over. Stitch felt different, The streaming canvas made the process feel much more collaborative and interactive. Watching designs appear in real time created a stronger sense of momentum and made iteration feel natural.
The ability to import references and then refine sections through in-place edits was probably my favorite part of the experience. Rather than constantly rewriting long prompts, I could focus on improving specific parts of the design while preserving everything that was already working.
As someone who enjoys experimenting with interface concepts, that workflow felt significantly closer to how designers and developers naturally think.
One feature I'd love to see expanded further is even more granular visual control directly within the canvas, allowing creators to tweak generated elements while maintaining the speed of the AI workflow.
Overall, Stitch helped shorten the distance between a rough idea and a polished interactive experience.
Live Prototype: https://8d2a7837-aura-living-interface-system.netlify.app/
Link to Stitch: https://stitch.withgoogle.com/projects/6682479894200923414
This is awesome
Challenges
View all
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
portfolioreview
The best portfolios tell a story, not just show a grid. Share yours for feedback.
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?