Dhruv Pokhriyal grew up in Delhi with a curiosity about machines that Tony Stark only made stronger. Computer science as an elective in 11th standard started a love of coding, which had to wait through board exams and JEE preparation. After cracking JEE Advanced, Pokhriyal started building in earnest and is now studying for a CS degree at IIT. PickMyVibe is the second product, after StudySnapAI.
Two years of full-stack development in college kept running into the same snag: hours lost choosing colours and fonts that matched the feeling of an app. PickMyVibe turns a mood such as “rainy neon Tokyo night” into five colours with fixed roles, two Google Font pairings and three CSS gradients, measures the contrast itself, and hands it all over as code. We spoke about not trusting the model’s contrast numbers, a launch that taught distribution the hard way, and why the apps that feel different will be the ones a person actually edited.
Key takeaways
- PickMyVibe turns a mood or phrase into a small theme: five colours with named roles, two Google Font pairings with a reason, and three CSS gradients.
- It throws away the model’s contrast figures and measures readability itself with chroma.js, labelling each pair AAA, AA, AA Large or Fail.
- Results copy straight into code as CSS variables, a Tailwind theme snippet or a Google Fonts import. There is no login and no Figma plugin.
- It is free and bootstrapped. The 12-month targets: about 1,000 people generating a vibe and a simple paid plan bringing in about $500 a month.
Before any of this
Startup.talk:Where did you grow up, and what was childhood like?
Pokhriyal:I grew up in Delhi, like usual I used to go to school and play with my friends, but I always had some form of love and curiosity toward machines, then later I was also inspired by Tony Stark which further increased my interest in machines. But for the most part my childhood was spending time with family and friends.
Startup.talk:What first got you into building things?
Pokhriyal:I took CS as an elective in 11th standard and fell in love with coding but then didn't do much in my 11th and 12th grade, since I had boards and my JEE preparation, but once I cracked JEE Advanced I started doing coding more and building more things, and that increased more in my IIT journey.
Startup.talk:What was your first job, and what did it teach you?
Pokhriyal:I haven't had any job yet, I am still pursuing my CS degree from IIT. But I did one more product before PickMyVibe, that was StudySnapAI and it taught me a lot, about how to work in sprints, ideation, design decisions, pricing decisions and more.
The road to PickMyVibe
Startup.talk:What made you start PickMyVibe?
Pokhriyal:What made me start PickMyVibe was since I've been doing Full Stack Development for the last 2 years in my college degree, I noticed a lot of time we get stuck on decisions like what color would match the vibe or finding a right color palette, cause a right and wrong color palette makes a lot of difference, that led me to PickMyVibe, since I wanted to simplify the things and better UI.
Startup.talk:What frustrated you about choosing colours and fonts for your own projects that convinced you to build PickMyVibe?
Pokhriyal:The frustrating part in choosing colors and fonts for projects was, often times most of the fonts and colors were generic and were not specific to the application and we had to spend hours finding the right fit that match the vibe we want.
Startup.talk:Why did you decide to start from a mood or vibe instead of from design tools?
Pokhriyal:I started from a mood or vibe instead of design tools is because as humans we tend to feel mood or vibe of something, so since other applications that I or any other user of PickMyVibe will be creating will be used by users, so it should convey the meaning they want to give and it sits right with color theory as well that different color gives different vibes.
Startup.talk:What did you learn from the first developers and creators who used PickMyVibe?
Pokhriyal:As the developer of PickMyVibe, what I learned during my journey of making PickMyVibe, was iterative design and rapid prototyping since I believe that is a very useful skill in today's era. Another thing I learned was to ship an SLC, while an MVP is good, SLC is what makes user want to come back to your website.
“while a MVP is good, SLC is what makes user want to come back to your website.”Dhruv Pokhriyal
The product
Startup.talk:Why the name PickMyVibe?
Pokhriyal:The name PickMyVibe represents, how I want the product to function, where the user can specify the vibe they want or the thing they want to convey and then the product generates the color palette and fonts which help them convey the vibe.
Startup.talk:How does PickMyVibe turn a phrase like “rainy neon Tokyo night” into palettes, typography and gradients?
Pokhriyal:You type a mood, or pick one like Neon Noir or Cottagecore. That phrase goes to a design-system prompt that asks the model for one structured result: a title and mood words, five colours with fixed jobs (Primary, Accent, Surface, Text, Highlight), two Google Font pairings with a short reason they belong together, and three CSS gradients. The model is told to treat the phrase as one atmosphere, so the browns of a café or the cyan and magenta of a rainy Tokyo night show up as a set, not as five unrelated swatches. The app then draws that JSON as palette cards, type pairings, and gradient strips you can copy.

Startup.talk:How do you make sure the colours and fonts it suggests work well together and stay readable?
Pokhriyal:Colour roles do the structural work. Text, Surface, and Primary are named jobs, so the palette is a small interface system rather than a bag of hex codes. After generation, the server ignores the model’s contrast numbers and recalculates three pairs with chroma.js: Text on Surface, Text on Primary, and Highlight on Surface. Each pair is labelled AAA (7:1 or higher), AA (4.5:1 or higher), AA Large (3:1 or higher), or Fail, and that score is shown on the result. Fonts are chosen as Google Font display and body pairs, each with a written rationale for why they suit the mood. The score is a check you can see before you ship. A failing pair is reported, and the palette is not silently rewritten.
Startup.talk:How do people take a PickMyVibe result into their own code or design tools?
Pokhriyal:Everything useful is a copy. A single hex, the whole palette, or a gradient’s linear-gradient(...). Each type pairing copies a Google Fonts @import plus font-family rules for headings and body. The export panel turns the same result into CSS custom properties (--primary, --surface, --gradient-1), a Tailwind theme.extend.colors snippet, or a plain role-and-hex list. You paste that into a stylesheet or Tailwind config and keep building. There is no Figma plugin or file download. The handoff is clipboard to code.
Startup.talk:Who gets the most value from PickMyVibe today: indie hackers, developers or content creators?
Pokhriyal:Indie hackers and developers. It was built for people who already know the feeling they want (“calm”, “cyberpunk”, “late-night café”) and get stuck turning that into colours, type, and gradients, especially without a designer on the project. The output is aimed at that moment: hex values, a closest Tailwind name, CSS variables, and a theme snippet you can drop into an app the same day. Content creators can use the palette and mood words as a direction, but the result is a theme for code, not a post template or a brand deck.
Startup.talk:What do you deliberately leave out of PickMyVibe?
Pokhriyal:A full design system. There is no spacing scale, radius, shadow, or component library, and no generated screens, logos, or illustrations. Five colours, two font pairs, and three gradients are a visual direction, not a UI kit. It also does not log you in, save a library of vibes, or sync to Figma. Readability is measured and shown. PickMyVibe does not auto-fix a palette that fails contrast, because the point is a fast, honest starting point you edit in your own project, not a finished brand.
Positioning
Startup.talk:Who is PickMyVibe for, and what job does it do for them? Describe the person who gets the most value.
Pokhriyal:PickMyVibe is for an indie hacker or developer who can ship the product and already knows the feeling they want, but gets stuck turning “rainy neon Tokyo night” or “calm Scandinavian app” into colours, type, and gradients. Their job is the first visual direction: five colours with roles, two font pairs, three CSS gradients, a real contrast score, and something they can paste into CSS or Tailwind the same day. The person who gets the most value is building alone, or without a designer, and needs the interface to feel intentional before they write the rest of the UI.
Startup.talk:How do you view your positioning relative to Coolors, Khroma, Adobe Color, Fontjoy and AI design tools like Uizard and v0? Is there another competitor you think of first?
Pokhriyal:Coolors, Khroma, and Adobe Color are colour tools. Fontjoy is a font-pairing tool. Uizard and v0 jump ahead and generate screens or components. PickMyVibe sits in the gap: one mood becomes a small theme, with colour roles, type, gradients, a measured contrast score, and a copy-paste handoff into code. It does not replace a colour wheel, and it does not lay out your app. The competitor I think of first is that gap itself. People either collect swatches, or they ask an AI to draw the whole interface. PickMyVibe is the step where the mood becomes a theme you can actually build from.
Startup.talk:What is your core moat in AI-assisted design direction?
Pokhriyal:The moat is the shape of the result, not a private model. The prompt locks the output to one atmosphere: five named colour jobs, two type pairs with a reason, and three gradients. The server then throws away the model’s contrast numbers and recomputes Text on Surface, Text on Primary, and Highlight on Surface with chroma.js, so the readability score is measured. The export is already code: CSS variables, a Tailwind colour extension, hex, and a Google Fonts import. Colour tools stop at swatches. UI generators skip this layer and draw screens. PickMyVibe owns the translation from a sentence to a theme you can ship.
Startup.talk:What are some of the use cases for PickMyVibe that are not obvious and can add a lot of value to users? List as many as possible.
Pokhriyal:Spinning up a Tailwind theme for a hackathon or weekend app before any mock exists.
Putting two moods in front of a co-founder (“warm café” versus “brutalist”) so the argument is about a palette, not an adjective.
Grabbing a hero gradient and a surface colour for a landing page the day you name the product.
Checking that a dark, moody palette can still hold body text, because Fail stays Fail.
Pairing fonts for a blog, docs site, or portfolio and copying the Google Fonts import straight in.
Giving a designer or a developer the same five hex values so the first commit is not default slate.
Re-skinning a side project when you are tired of the current look, without redrawing every screen.
Setting one direction for a Product Hunt page, a pitch deck, and the app, from the same phrase.
Teaching yourself WCAG by reading AAA, AA, and Fail on a palette you just described.
Theming a game jam HUD, an event page, or a music project from an atmosphere instead of a logo.
Writing the intended mood into a pull request or a design note, with colours attached.
Trying “trustworthy fintech” and “playful fintech” as two generations, then keeping the one whose text-on-surface score you can live with.
Matching a newsletter, slide header, or social banner to the same Primary, Surface, and Accent as the product.
Starting a client conversation from their words (“late-night studio”, “desert minimal”) and showing a direction in the meeting.
Building the personal site you have been postponing because you could not choose a type pair.
Startup.talk:What is the one detail in PickMyVibe you are proudest of?
Pokhriyal:The interface itself. The line “Turn vibes into beautiful design systems.” focuses one word at a time: that word stays sharp, the others blur, and a frame moves onto the active word. A soft blob trails the cursor. The menu opens as staggered colour layers, not a plain list. An aurora sits behind the whole page. I wanted the product to feel like a vibe before anyone typed one.
Launch
Startup.talk:What did you learn from your launch?
Pokhriyal:Shipping the listing was the easy part. Product Hunt did not feature PickMyVibe, so the page never became a real conversation: a handful of upvotes, and the only comment was my own. I learned that a launch is a distribution problem. The maker note explained why I built it, and that still left people with nothing to react to. The useful responses came from sharing it myself. People answered the feeling of the product and the problem, which told me the next launch needs a sharper ask, a live example of a phrase turning into a theme, and an audience I bring, rather than a hope that the homepage will do it.
“I learned that a launch is a distribution problem.”Dhruv Pokhriyal
Startup.talk:Which feedback from the launch surprised you?
Pokhriyal:I asked for brutal honesty and barely got any. I expected people to pick apart the colours, the fonts, or the things the product does not do yet. The comments that did show up were short. On LinkedIn, one person said it looks lovely, and another said this is something designing needs. Nobody sent a feature list. The surprise was how little criticism there was, and that the first reaction was the look of the product and the problem it is trying to solve.
The hard part
Startup.talk:How many people use PickMyVibe today?
Pokhriyal:I do not have a public user count. PickMyVibe is a free web app, and the Product Hunt launch was small, with the conversation happening more on LinkedIn than on the listing. Usage is tracked in PostHog per generation, copy, and export, so I can see activity, but I am not going to put a made-up monthly number on this form.
Startup.talk:What is a challenging time you had building PickMyVibe?
Pokhriyal:The hard part was trust. The model will happily return a contrast ratio, and that number can be wrong while looking precise. I stopped using it. After the palette comes back, the server throws those ratios away and measures three pairs with chroma.js: Text on Surface, Text on Primary, and Highlight on Surface. Building the page had its own snag. The aurora background is WebGL, so it cannot render on the server. It has to load only in the browser, or the app breaks on first paint.
“The model will happily return a contrast ratio, and that number can be wrong while looking precise.”Dhruv Pokhriyal
Startup.talk:Who believed in you before there was anything to show? What did they say or do?
Pokhriyal:My family believed in me, they supported me that in case I need any help they are ready to back me up. My dad funded the GPT tokens and also checked the design of my website once I completed it.

Startup.talk:How do you keep AI-generated design systems original rather than generic?
Pokhriyal:A result stays tied to the phrase because the prompt has to return one set: five colour roles, two font pairs with a reason, and three gradients. "Rainy neon Tokyo night" and "warm indie café" share that shape. The colours, type, and gradients are what change. I do not yet reject a palette that comes back generic. Font names are not checked against Google Fonts, and the Tailwind name is the model's guess. That is the part I am going to improve. Next I want a check after generation: measure the real contrast, confirm the fonts exist, and regenerate when the palette ignores the phrase and falls back to a default look.
What it is all for
Startup.talk:What would make you say PickMyVibe has succeeded?
Pokhriyal:A developer describes a feeling, pastes the result into a real project, and ships with it. The colours, type, and gradients still match the phrase a week later, and the text stays readable because the contrast score was measured. They come back for the next product instead of opening a blank Tailwind config and guessing.
Startup.talk:What will design look like when anyone can describe a feeling and get a finished visual identity?
Pokhriyal:Getting a first look will be easy. You describe a feeling and get the colours, fonts, and gradients right away. The real work will be choosing what fits the product and fixing what does not. A lot of apps will start from a similar draft, so the ones that feel different will be the ones a person actually edited.
“the ones that feel different will be the ones a person actually edited.”Dhruv Pokhriyal
The road ahead
Startup.talk:What are your milestones for the next 12 months, in users and revenue?
Pokhriyal:Over the next 12 months, about 1,000 people who have generated a vibe, and about 100 who copied a theme into a real project. Revenue stays small on purpose. PickMyVibe is free today. Within those 12 months I want a simple paid plan for saving vibes and cleaner exports, with a target of about $500 a month.
Startup.talk:And for the next 24 months?
Pokhriyal:About 5,000 people who have generated a vibe, and about 200 paying users. Revenue target is about $2,000 a month. The product should remember a project’s vibe, check the fonts, and regenerate a palette that ignores the phrase.
Startup.talk:Where do you want PickMyVibe to be in five years?
Pokhriyal:The first place a builder goes when they know the feeling and need a theme they can ship. They type a sentence, get colours, type, and gradients they trust, and paste them into their own product. PickMyVibe holds the starting direction. The person still decides what stays.
Startup.talk:What are your hopes and fears for a post AGI world?
Pokhriyal:One fear that I have for post-AGI would be that creativity would decrease and people would start using generic templates, but I still have hope that people would always want their product to give a certain kind of vibe so PickMyVibe would still thrive in that.
This interview was conducted in writing and has been edited lightly for length and clarity.
About PickMyVibe
PickMyVibe is an aesthetic copilot built by Dhruv Pokhriyal in Delhi. It turns a mood, vibe or phrase into a small design system: five colours with named roles, two Google Font pairings and three CSS gradients, with contrast measured by the app and exports as CSS variables, a Tailwind theme snippet or a Google Fonts import. It is free to use.
| Attribute | Verified Detail |
|---|---|
| Company | PickMyVibe |
| Founder(s) | Dhruv Pokhriyal |
| Country | India |
| Grew up | Delhi, India |
| Studying | Computer science, IIT |
| Before | Built StudySnapAI |
| Funding | Bootstrapped |
| Official Website | PickMyVibe |
More founder interviews
For founders
Building something? Want to share your journey?
Tell us how it started, what nearly broke it and where it is headed. We will turn it into a story like this one.
Share your journeyFor readers
Want to read more human stories behind the startups?
New founder interviews and stories, as they are published.
Follow us on LinkedIn
VoCalm
Skillbox
VedIX
Zyaku