In partnership with

The Daily Newsletter for Intellectually Curious Readers

If you're frustrated by one-sided reporting, our 5-minute newsletter is the missing piece. We sift through 100+ sources to bring you comprehensive, unbiased news—free from political agendas. Stay informed with factual coverage on the topics that matter.

Join for free today!

Hey, I’m Kushagra.

AI design isn’t just code and math—it’s about trust, vibes, and real impact. But how do you make it work step by step?

Most MVPs? All hype, no real check. Teams rush, launch fast, then—silence.

From winning trust to scaling smart, this guide keeps it clear, real, and hype.

How to Build Cool GenAI Stuff

GenAI is everywhere—new drops hit the news daily. But dope GenAI needs a dope design. Before you start, know this: GenAI is not old-school AI.

What is the big thing with GenAI? It’s wild. Same input, different output. It makes words, pics, and code—not just picks or tags. That shift? It's huge for design.

Each step comes with key Qs to help you think before you build. It's not the full list, but it's a solid start.

These steps hit hard ‘cause they make you see the whole vibe—not just what’s on the screen. GenAI ain’t just looks—it’s deep, messy, and full of wild human-AI moves.

This guide helps you think past what you see—so your design ain’t just useful but next-level. If that sounds dope, let’s jump in.

Know What GenAI Can (and Can’t) Do

Most AI stuff you’ll work on won’t just drop raw GPT, Claude, or Llama. It’ll be tweaked for a real use case. That means the AI team will fine-tune it, and you gotta know what’s up—what it can do, what it can’t. Your whole design rides on that.

You don’t need to be an AI nerd (unless you want to), but you do need the right mental map to see how this tech shapes the user's vibe.

Here are some key Qs to get your head right: ⬇️

Think big—what kind of stuff can this AI take in and spit out? Text, pics, sound, vids? Know the mode, and shape the flow.

This matters big for design—if the AI lags, what’s the wait vibe? ⏳ It also shapes how users click, tap, or chat with it. Keep it smooth.

This shows what the AI can do and say. Wanna go deep? Look up RAG and fine-tune—that’s how it gets smart.

This decides how much the AI remembers. Small brain? Users gotta repeat a lot. Big brain? Long chats, big code drops. 

Know What Users Want

Can’t say this loud enough—this is key.  Most designers map the old flow, find the pain, and fix it. That’s cool, but with GenAI, we gotta think differently. I can’t emphasize this have to think differently

Instead of just tweaking what’s there, ask these Qs: ⬇️

The Real Job to Be Done 

It’s not just about what users say they need—it’s about why they need it. The job to be done is the big goal, not just the tool they pick.

Like, a user says, "I need a car to get to work." But the real job? "Move fast from A to B." That means cars, bikes, buses—all options open. For example, might say is and are 

Think big, not just what’s now. That’s how you break free and build next-level stuff.

Don’t box yourself in with just today’s personas. Tech’s moving fast—today’s roles might not even exist tomorrow. Always start with the big goal—the job to be done—and keep those personas in mind. ✨

Always think big about why certain personas exist. Is it ‘cause tech can’t do more? Or humans are just limited? If a persona’s tied to rules or laws, it’s probably staying. But if it’s ‘cause of tech limits? That might change soon. 💥

How Humans and AI Work Together

We’ve all seen AI magic—and we’ve felt the letdown. When AI hits, it’s fire, but when it misses, it’s frustrating. Depending on how confident the AI is, a human’s gotta step in somewhere. This step is where you decide how AI and humans will team up. There are 3 main ways they vibe:

1. Hands-off:
   Humans give AI a task, then just chill while the AI does its thing. Afterward, the human checks what’s done. Example: A marketing manager tells AI the goal, and AI makes the campaign—no fuss. It’s quick but basic and great for tasks that don’t need a lot of creativity.

2. Co-create:
   Humans and AI work side-by-side, sharing the load. They both help build the result, bouncing ideas off each other. For the marketing campaign, the manager writes a draft, and AI gives options or makes tweaks. It’s creative and interactive, perfect for projects where you need ideas to flow.side by sideThis process isideas need

3. Complementary:
   AI and humans do their own thing separately, then bring it together. Each side creates its version, then merges to make the best mix. The manager might work on social media while AI handles the email marketing. The combo gives you the best of both worlds.

Each of these models has its vibe, made for different needs—control, complexity, or creativity. To nail down how humans and AI should work, ask yourself:

Think about if the work is something users have to do or if it adds value. If it’s more of a chore, the hands-off model is a good fit. Like, keeping code up-to-date? Yeah, that’s a perfect hands-off job. But for stuff that needs creativity, like adding cool new features, a co-create vibe works way better.

You gotta get this—no matter how we wish, timelines and deadlines run the show. We need to do stuff fast. So, it’s key to know what tech can back us up.

What I found amazing this week

ASUS reinvents the mouse—navigation meets aromatherapy. Future of work?🌿

This track gave me a serious boost—check out ‘Games’ by The Strokes🎵

What’s the real reason POOLS is designed like this?✨