2026-02-27 08:00:00
Twitter is still undefeated for hiring.
We just closed our Founding Designer because he saw a tweet and DM’d. Now we’re hiring developers the same way. One of the best recruiters in India told me his conversion rate: 1 hire for every 50 conversations. For us it’s probably 1 in 100. We want to keep the team small and only work with the best.
But this post isn’t about the open role. It’s about how we think about hiring in 2026. Because I think the rules have changed and most companies haven’t caught up.
Let me start with a story.
6 months ago I was chatting with a friend. One of the sharpest devs I know. CTO-1 at his company. Someone who’d be running the entire engineering org at a startup eventually. Back then he was thinking about his career path. Manage more people, become a CTO, get more equity. The usual career ladder climbing stuff.
I talked to him again 2 weeks ago.
He has removed all reporting lines. Zero direct reports. All he does now is manage agents. Junior devs just slow him down. He doesn’t think layers of management will exist in a few years.
He went to his founders and said “let me just build with agents full time.”
He told me he’s scared too. He doesn’t know what happens to software engineering from here. So his short term play is to operate at the highest level of abstraction possible and let agents handle everything underneath.
This is one of the best devs/EMs I’ve worked with in my career. And he’s restructuring his entire role around AI.
If the best devs I know are already worried about the future. If companies like Block are laying off 40% of their workforce. Then what kind of people should an early stage startup even look for?
Here’s how I think about it.
There’s no “1 month KT” to ramp up anymore. You can have an agent explain the full context of a codebase in an afternoon. Most dev docs now let you chat with them. Cloudflare rebuilt Next.js in a week. (Okay, experimental. But agentic coding only gets better from here.)
The things that used to matter, years of experience in a specific stack, deep familiarity with a codebase, knowing where all the bodies are buried, those things are getting commoditized fast. An agent can get you 80% of that context in a day.
So what actually matters now? I keep coming back to three things.
High slope. How fast can you learn? Not what you already know. How quickly can you pick up something new and run with it? The tools are changing every few months. The person who learned React in 2018 and stopped there is less useful than someone who picked up three new frameworks this year just because they were curious. High energy. Do you bring the juice? This one sounds soft but it’s not. Startups need optimists. There’s a lot of shit hitting the fan every day. I need people who walk in and raise the energy in the room. Not fake positivity. Real momentum. The kind of person who sees a problem and gets excited about solving it instead of listing reasons it’s hard. High taste. Do you know what good looks like? Are you obsessed with craft? Do you have opinions about what makes great software great? Taste is one of those things that’s hard to interview for but impossible to miss when someone has it.
And agency.
We recently hired our founding designer. Here’s what stood out.
I told him upfront that we will hire only a full-stack designer. Product design. Visual design. Even illustrations sometimes. The craft isn’t in pushing pixels anymore. The moat is taste.
I give open ended assignments to test people. To him, I gave the task to think through our brand and homepage. He came back with a dozen different directions. Different IA, different themes. Working on details no one would expect from a take home assignment. He kept following up. Kept trying to understand what we’re building.
That alone was impressive. But what sealed it was the follow-up chats.
He’d tried every AI design tool out there. He also follows all my favourite designers. Vercel’s team, Linear’s team.
Every cool landing page I had bookmarked, he had them in his bookmarks too.
I think about developer hiring the same way now.
The job has changed. You’re not just writing code anymore. You’re managing a fleet of agents while having the technical depth to architect systems. You need to operate at every level of abstraction. System design one day, debugging agent output the next.
The developers I want to work with are already using AI coding tools. Claude Code, Cursor. Codex.
They’re excited about where this is going, not threatened by it. They learn fast. They bring energy and optimism. They make the people around them better. They have taste. They know what great software feels like because they’ve studied the best.
I don’t think this is just how we hire. I think this is where hiring is going for most startups.
The resume, the years of experience, the specific tech stack. All of that matters less every month.
Slope. Energy. Taste. Agency. That’s all that matters.
2026-02-17 08:00:00
Before I worked with him, I had met good designers. People who are the absolute best when it comes to interaction design, the kind who would sit and think through every possible way you can interact with a component on screen. Then there are people who are brilliant at visual stuff but have no idea about basic interactions. And of course, there are people really good at the product design stuff: thinking holistically about why a feature is built, what it does, what kind of things you should ship. Almost working like a pseudo PM. If you look at fidelity, there are different levels you can work with. Low, mid, high. I’ve worked with people who are now designers at tier-one companies who refuse to turn their mid-fidelity wireframes into visual designs. They rely on a design system and expect some visual designer to turn their concepts into reality. They don’t want to worry about margin and padding.
But Vaibhav? He’s the guy who would be jamming with you on a whiteboard on very complex user problems. From concept to different levels of fidelity to pulling ideas from completely different domains to shape the product.
I have a similar interest in design. I’ve read dozens of books on product design, branding. I spend time on Midjourney figuring out abstract ideas. I like sitting on Figma doing random posters. But I am probably 60% of everything. And you can’t get to a tier-one product by being 60%.
But I know what needs to be done. I’ll sit on Google Stitch and iterate multiple designs for a product card. I’ll use Claude Code and iterate on a dozen different ideas for a homepage. Style transfer is pretty easy with AI now. In today’s age, I don’t think you really need to sit on Figma and grind out every detail. It’s very easy to move between fidelity levels. You can jam with Claude, build a skill around how to think through branding, open up the branding guidelines of five great companies: Wise, Square, Klarna, whatever. Figure out how they do typography, how they’ve done colors, how they think through the brand. You can study how other great brands have been built, what they stand for, and come up with what you want the user to feel when they use your app and what kind of users you want to attract. But even though I have the ideas, it is very hard for me, as someone who is probably not even 60% when it comes to branding, to produce the actual artifacts. BTW Gojek had one of the best brand books ever. Kudos to Abhinit and gang who came up with it.
When I talk to designers now, they are rigid. They want to do only interaction design. They don’t want to think about working on, say, your logo because they think it’s a different skill, a different craft. And sure, they should focus on being spiky in one dimension. But the reality is that roles are compressing. Even though you need 99th-percentile talent and skill across all these things, in the early stage, you want people who do everything.
Forget early stage when we were a public company, Vaibhav was working on killer decks for the product features we shipped.
And that is what I really miss.
I miss working with someone who would be excited about jamming on an idea late at night. Someone who has the taste and agency to turn it into a wireframe where you can jam more on a specific direction. Then the taste and the craft to bring it to life. And more than all of that. Someone who doesn’t think creating a deck to highlight a feature is beneath them because they’re now “Head of Design.”
There are very, very few people like that. Trust me, I have tried. The only other designer who came close was another designer who was even better in terms of craft. But for some reason, he was insanely hard to work with. One of those mega-talented people (with extremely bad people skills) who are just difficult to collaborate with. But Vaibhav was a gem of a person to be around too. Kind. Happy to give more than he takes. So it becomes very hard to compromise.
So yeah. This long rant is really about one thing: we need a tier-one designer. We raised one of the biggest seed rounds in India last year. Backed by tier-one VCs. We’re building a product in the prediction market space. Initially remote, but happy to discuss relocation to Dubai. We’re willing to move for the best talent. Happy to pay the best salary — but it depends on where you are on the experience curve, the skill curve, and what you bring to the table. The more skills you have, the more willing you are to push yourself, the better. We grind a lot. So we expect you to also.
That’s why this is a cultural fit only if you’re the kind of person who likes trying out the latest design tools the day they launch. Someone who keeps a mood board for typography, designs, colors. Who loves spending time on Cosmos. The kind of person who tried Variant AI the day it dropped. Not the kind of person who rates themselves 5 out of 5 on interaction design with no proof of work to back it up.
And definitely not someone who thinks building a deck or spending 15 minutes on Figma to create a hiring post is beneath them.
2025-11-28 08:00:00
Since I started integrating AI deeply into my workflow, my baseline for quality has shifted dramatically.
A few days ago I was iterating on a single UI card. I kept taking screenshots, dropping them into NotebookLM, feeding it references (books, patterns, screenshots), and asking, “What’s missing? What would a good designer point out?” I did the same loop across a few other tools too. My goal was to reach 70-80% of a professional designer’s output.
Most models caught things I missed. And the more context I gave, the sharper the responses.
Ryo from Cursor recently wrote that in the future, teams will be smaller, with more generalists.
If that is true, then what happens to the average specialist?
If I can generate a better artefact with just a few prompts than what a designer creates, I’m going to be disappointed with that person’s output.
The baseline is higher than it has ever been.
AI has fundamentally broken the “average” tier of creative work. Average design and average landing pages don’t hit the same, because “average” is something a model can do by default.
So, where is the alpha?
I recently saw Alter Magazine’s website. The hero section. The typography, all the imagery. The website just had soul.
It felt like it was created by someone who really cared. Yes, AI can generate. It can remix. It can get you to “pretty good” fast. But the real alpha is going beyond what the average model would produce: making choices that are surprising, but clearly intentional.
That is what models can’t replicate yet. The alpha is in taste, curiosity, and the ability to craft something that feels human in a sea of AI content.
2025-11-20 08:00:00
“Oh Manas, why do you keep posting these screenshot essays on twitter? Just write a normal long post, it is better for distribution.”
No, you do not understand. I am a Figma Pro user. Like every middle class Indian who has paid for a tool, I must extract full value from my subscription.
That is why:
Thank you for your attention to this matter.
2025-11-19 08:00:00
Don’t use NotebookLM only to consume others’ ideas. Use it to clarify and compress your own thinking so that you can present it better.
A lot of people use NotebookLM only for consumption. Drop in a paper, get a summary, maybe a video or slides.
But where NotebookLM really shines is creation.
A friend showed me his workflow around using NotebookLM to produce the first version of anything he needs to explain. He’ll drop in his notes, drafts, and prototypes, and NotebookLM gives him a tight five mins video brief that he refines to come up with the final output that he presents to his CXO.
I have started doing the same. I had a long manifesto for my startup and wanted to turn it into a short elevator pitch. NotebookLM compressed it perfectly and gave me a clean base to iterate from.
2025-11-18 08:00:00
The fastest way to get started with a new project, especially if you want to copy an existing product and iterate on top of it:
Use the Magicpath Chrome extension With this extension, you can select an element on a webpage and copy-paste it into Magicpath. It will automatically generate a React component that you can use. However, there’s a limitation: while you can view the component, you can’t export the code unless you pay for Magicpath. The output is pretty good. Trust me, I’ve tried many vibecoding tools. It’s still hard for them to visit a website and copy either the whole page or the individual components. And yes, I know Cursor can now see your browser, it still doesn’t work.
Figma Make -> Figma Design -> Claude Code Another approach is to take screenshots of the components you want. Then, go to FigmaMake and ask it to create the component from the screenshot. You can one click export the component from Figma Make to Figma Design. If you’re a designer who prefers working in Figma instead of dealing with code, this method gives you more control. You can then refine the component directly within the Figma design environment. Once you’ve finalised the design and are ready to start building, you can use the Figma MCP + Claude Code client to build a working prototype.