RunEvr
Agentic project management environment for creatives
Start new thread

🎨 Do solo makers need a UI/UX designer?

by
•

For solo makers, wearing every hat is almost unavoidable - developer, marketer, support, product… and often designer too.

AI makes this even easier now. You can generate UI pretty quickly, but a lot of products end up relying on consistent, polished AI-generated solutions rather than something truly unique or thoughtfully designed for their specific product.

So I’m curious:

Do you design the UI/UX yourself when building solo?

Or do you think bringing in a designer (even for a few key screens) is worth it?

And if you’re using AI for design, do you feel it’s enough to get you to a good product experience?

Would love to hear how other makers approach it.

3.3K views

Add a comment

Replies

Best

For me onboarding is where i would spend the design budget. if user dont understand the product in the first few minutes, polishing the dashboard probably wont save it.

 Agree 100%! Onboarding is key.
What strategies do you use - interactive onboarding, tutorials, or making the initial experience very intuitive?

I have found that polished screens can still create confusing experiences. Testing with real users matters more than making every screen beautiful.

 In the case of a UI/UX designer, it isn't more about ,,beautiful and polished,, but about user experience first.

A professional designer combines both - beautiful and intuitive design. And it’s all about micro-decisions like button placement, color, spacing, and so on.
I think even with testing results, a designer can interpret them and make wiser changes based on experience.

I think AI is great for getting a solid starting point, especially for solo makers. It can help you move much faster with layouts, ideas, and iterations.

But for the final product experience, I still think a designer can add a lot of value. Even bringing one in for a few key screens, user flows, or a design system can make the product feel much more polished and intentional.

So for me, it’s less about AI vs. designer and more about using AI for speed and a designer for the details that make the experience feel unique.

Great wording! Are you currently building a product? What stage are you at?

I would say you don't need one for the start because Claude design can give you a very good starting UI/UX layout. Especially if you have a product background or a basic understanding of UI/UX, you can get away with it for a very long time until you get to a proper point where you can afford a really good UI/UX designer. Even then you can just get one superstar who can use the tools to find the best solutions for you. Especially if you have data, you can reuse that data to fit into your AI brain for UI/UX to track actual user behavior to build an actual better product

I agree with you, especially for the early stage. But do you think Claude Design or other AI tools can really give your product a unique approach? AI models rely heavily on existing best design patterns and proven solutions, which is great for getting a strong starting point. But when a product designer spends time understanding your product, your thoughts, and the specific problems you’re solving, they can come up with something new and add it to those existing design solutions. That combination of proven patterns + human interpretation is where I think the real uniqueness comes from.

 of course they have their limits, but I think they are really good enough to get you to a stage when you need to worry about the uniqueness. First priority is usability imho

totally agree!

Six years as a graphic designer before I started building Merzify solo, so I'm on the opposite side of the usual problem in this thread. I don't need AI to generate UI, I need AI to help me stop overdesigning it.

Design background actually slows me down sometimes. I'll spend real time polishing a screen nobody's asked for yet instead of shipping the ugly version and finding out if the feature even matters. The instinct to make it look right before it's proven right is a real trap.

Where AI genuinely helps me: not generating the design, but checking my own bias. I'll ask it to poke holes in a layout decision before I commit engineering time to it. Different use than most people in this thread are describing.

My honest answer to your actual question: if you don't have a design background, bring someone in for the handful of screens people actually judge you on in the first ten seconds, onboarding, the core result screen, whatever your "moment of truth" screen is. AI-generated UI is fine for internal tools, admin panels, anything the user doesn't emotionally react to. It gets noticeably weaker the moment the screen needs to earn trust, and that's usually the screen that matters most.

 Really enjoyed reading your comment! Thanks for sharing your experience from a designer’s point of view, especially the part about AI helping you challenge your own design bias.

What problem does Merzify solve? And are you planning to launch it here on PH?

 Appreciate that. Merzify checks whether an Amazon product is actually safe to buy before you spend money on it. Reviews get faked at scale now, AI generated, coordinated, timed in bursts, and around 30% of Amazon reviews are fake at this point. Most shoppers have no way to tell a real five star rating from a manufactured one.


Merzify looks at the reviews, the seller, and the price together, then gives one straight verdict instead of a score to interpret yourself. Safe to Buy, Proceed with Caution, Uncertain, or Trust Cannot Be Established. If the data isn't solid enough to answer honestly, it says so instead of guessing, that's the rule I care most about not breaking.


Still pre launch, but yes, planning to launch properly here once it's ready. Will probably be back in this thread asking you the same design question once I'm closer.

 I really like the idea behind Merzify!

Do you already have an estimate of how accurate its verdicts are? Like, what percentage of the time does it correctly identify whether a product is trustworthy to buy?

I’ve been on both sides of this. When you’re solo, you have to design, there’s no way around it. AI helps you get to something usable faster, but most of what it generates sits in the “acceptable” zone, not “intentional.” The biggest gap I see isn’t visual polish, it’s thinking. Flows, edge cases, states, what happens when things break. That’s where products start to feel generic.


Personally, I’d design the core flows myself early on, then bring in a designer later once there’s some clarity and real usage. Even a few hours of focused input can shift the product from “works” to “feels right.” AI is a great accelerator, but it still needs taste and judgment to land well.

 Liked the thought about the “gap in thinking.” That’s what’s missing in the consistently polished designs of AI today.
What problem does your product solve?

 That gap is exactly what we’re focused on with CambrianEdge.ai


Most tools give you powerful outputs, but they’re still fragmented. You get pieces, not a cohesive flow. The hard part isn’t generating UI or content, it’s connecting research, thinking, creation, and execution into something that actually works end to end. We’re trying to make AI feel less like a set of tools and more like a system that helps you move from idea to outcome without losing context along the way.

 Wow! Impressive. Just signed up - I’ll give it a try, and if I have any feedback, I’ll come back with it.

Personally i found having check ins with friends during the build helped alot. I literally hand them the phone and give no context, no words, just observe. That helped me alot with UI/UX with navigation and the app explaining itself essentially!

 Yes, this is a really cool approach to understanding how your interface actually works. What are you building? Is it a mobile app?

It helps to consult a designer but you can use AI to help you without. AI is good at making UI look “pretty” but that’s never been what UX design or UI is actually about and pretty UI doesnt make usability better . UX helps someone complete a task without having to think, and designers tend to understand common design “conventions” that lead to an easier understanding. AI is good at UX conventions only when you describe the context and problem you’re trying to solve. Most people describe what they want in features (this will get the worst design result because AI has no context of the specific problem) so I would say if you know a designer get them to have a look. But it’s not always always necessary. I can say that as a UX designer with over 10+ years experience.

 I think AI works well mainly because it’s in your hands - a designer with 10+ years of experience =)

I can definitely say that experienced designers can squeeze much more out of AI than someone without a design background. Wouldn’t you agree?

I used to focus primarily on UI/UX, but I no longer see how I can add value by doing the work I used to do—taking ideas from the product team, turning them into mockups or prototypes, and building the React components with state wiring after long approval cycles. All of that can now be done in minutes with Stitch.

When I built Handysuite, the initial UI/UX was generated by Stitch based on my prompt, and it was better than what I would have designed myself. After building an MVP using those designs with Cursor, it looked great. However, when I actually started using it, I ran into many user experience issues. At that point, UI/UX expertise became essential. Testing the product, evaluating how it feels, and crafting solutions to refine the experience cannot be done with AI agents alone, in my experience.

 Yes! AI tools are a real treasure when they’re in a designer’s hands! I can confidently say that designers, with their expertise, can do real magic with them.

I’ve never tried Stitch before - definitely going to give it a shot!

not until someone has said they'd pay. bad design loses you people who already wanted the thing. no design problem explains zero interest in the thing itself, and solo makers usually spend the design money before they know which one they have. adana, of the products you've seen struggle, how many were actually a polish problem versus nobody wanting it?

 Yes, I share your opinion!

The main thing is to have a product that actually solves a real pain, in any way possible. But once you’re confident in the idea, I think good UI and a smooth user experience are a must-have.

 agreed, and the phrase doing all the work in your answer is "once you're confident in the idea". most people declare that far too early, usually on the strength of people being nice to them.

the test i use is whether anyone has given up something to get it. money, an email they actually check, a slot in their week. before one of those has happened, polish is being funded by the founder's own optimism.

after that line i think you are more right than i was. bad UI at that point is not cosmetic, it is a refund. someone who had already decided they wanted the thing leaves because a screen confused them, and that is the most expensive churn there is because you already paid to get them there.

what made you draw the line where you did? feels like you have watched a specific product get this exact thing wrong.

 I really like reading your points of view! They’re interesting enough that I actually read them twice =)

And the answer to your last question is pretty simple: I’m the UI/UX half of our founding team =D That’s probably why I care so much about it.

First
Previous
•••
345
•••
Next