Barveen Kumar

VENI. VIDI. VICI.

@dara_venture OMG. feels like most of this is meant for Kuralit project.

If I had to raise pre revenue here's what I would do...... Lead with a sharp, almost contrarian insight. Your unique point of view about the market is your most valuable asset pre revenue. Investors fund "this person has seen something the market hasn't yet." If your thesis could come out of any investor's own mouth, you have absolutely nothing. "everyone believes X, but X is wrong, and here's why now is the moment." At minimum, you need a landing page that converts cold traffic, a REAL waitlist with quality signups, signed LOIs or paid pilots, design partners who've committed engineering time, and best of all.......usage and retention from a live product. Retention curves that flatten (people come back week after week) are the single most persuasive non revenue signal that exists. Investors at this stage are betting on your rate of improvement and not necessarily your current state. Show them where you were 8 weeks ago versus now. Shipping speed, learning speed, how fast can you turn a customer conversation into a product change. Two founders with identical products but different slopes get wildly different outcomes. FACT Run a tight, compressed process to create competition. Don't drip out meetings over three months. Cluster them. Scarcity and FOMO are real fundraising mechanics......investors move quickest when they believe someone else is about to move before them.......aim to have multiple parties looking in the same two week window. (this must be genuine) Show you do a lot with a little. Capital efficiency is itself a signal of judgment. "We built this and got these signals on $40k and two people" A couple of these fork hard by sector...... deep tech / biotech raises almost entirely on team and technical de-risking. Consumer raises on growth and retention curves, B2B SaaS on pipeline and design partners. Being pre revenue is difficult......but there are effective ways to approach it.

@shcallaway learnt this the hard way. focusing on improving the craft along the way for @KuralitHQ

Fundraising at the seed and pre-seed stages is about storytelling. - Who are you? - What are you building? - Why are you building it? - What's different? - Why are you the right person to do this? - Why is it going to get really big? - What has changed? One of the big mistakes that I see first-time founders make is over-fixating on revenue instead of trying to create the most compelling story possible. It's much more important to get a customer that LOVES your product and validates your hypotheses about the world and the future than it is to demonstrate revenue growth - especially if that growth involves small absolute values over a small time window. Investors will fund you based on the quality and credibility of your story. Nothing else matters at this stage.

@garrytan maybe we should treat our @KuralitHQ's engine(brain) as multi-level (layer) cognition @itskeyans

Everyone building AI agents is focusing on building the prefrontal cortex. Planning. Reasoning. Multi-step chains. There's value here. CEO-stuff. But also, a reframe: there is value in building the cerebellum. It's offloading boring tasks into reflex so the complex thought can focus. Your mortgage gets paid by a standing order, not a committee. The things that are not fun, not interesting, but have to be done? Done. Most agent frameworks will fail because they treat all cognition as high cognition. The winners will nail the boring stuff first.

@vanshyadav1408 @ycombinator a good side hustle right?

A @ycombinator W26 founder proposed me $10k for a referral😭. Like dude, why?😭

@dara_venture tried the same on our YC application video. let's see.

Took a pre-revenue pitch this morning. Mid convo the founder goes: "you're probably wondering why the f*ck I'm the guy to build this." Then he told me. And crushed it. He killed the objection before I could even raise it. Don't wait for the doubt......name it out loud, then bury it.

@GaddipatiHarsha So, it's all about understanding my persona to handle the incoming conversation(s)? Or can I also use it to outreach (better use case for slashy I think). Also, possible to maintain inbox zero?

Excited to announce the first email client that actually learns. Slashy builds deep user personas to draft replies, label emails, and flag missed follow-ups. Users who switch from Superhuman save an extra hour a day. Try it at slashy.com

@tsvillain @adxtyahq @itskeyans check this out

@adxtyahq Everyone should have this free Eval Cheatsheet available on firsttoken.dev

People call themselves AI engineers after a LangChain/LangGraph tutorial If I told them how many prompt templates, tools, retrieval pipelines, and agent loops I deal with daily, they might cry. One moderate edit run 👇

@tsvillain We need to talk

After months away, I’m back on pglens.org! I never lost interest, the early “no-framework” decision just stopped scaling with the ideas I had. The new goal: turn pglens from a simple database viewer into a full no-code database tool for backend engineers. One that cuts ~80% of SQL writing with a clean UI-first approach. First step: migrating the entire stack. That alone will unblock everything I want to build next. The original vision is unchanged: a lightweight local database tool you install with a single curl command on Linux, macOS, or Windows. Zero setup headaches. Lots to build. Sharing progress here to keep myself accountable. #buildInPublic

We spent years blocking bots from accessing our software. Now we're redesigning that same software to make it easier for AI agents to use. Times change fast.

we just kicked off s4 last night. -300 online. -20 in-person. founders booking their flights to seoul. 4 weeks to build! welcome to korea everyone 🇰🇷 ty to @arlanr for the q&a session!

Anyone got access to @Swiggy's builder club?

@wh0sumit Sent you a DM while back on WhatsApp bro. Welcome back.

i just love talking with people man. back to replying, giving feedback in communities, and responding to DMs again. honestly missed this for the last few months now. sorry for the inconvenience from my side, but back with much more energy again 👊 if you reached out to me months ago and i never replied back, please text again. happy to chat, review your work, and connect.

@tsvillain Exactly.

we all just complain about chrome eating a lot of memory but there is no other alternative at the moment it feels like everything is just a wrapper

Day 1 of Zerobase Cohort 4 and the energy was insane with @samthekorean A bunch of internet people from everywhere, all jumping into a YouTube live to build crazy things together for the next few weeks. I’ll be building Kuralit during the cohort. Trying to improve how apps interact with humans through adaptive AI interfaces instead of static screens and endless taps. This already feels less like a “program” and more like the internet accidentally forming a hacker house. Let’s build ✌️

@JakeManger Got it. Thanks for the clarification.

@barveen_kumar_ Mainly because it’s easier on macOS and weirdly enough, mac users pay more for software (just from my past products) This app will be coming to windows soon though! (Requires a bit of a rewrite but it’s coming!)

why people makes apps (mostly screen recording software) for mac only? better paying customer segment? what am i missing here?

@thedankoe’s “5D thinking” is interesting because it pushes people beyond surface-level thinking. Most people react to events. Very few think in systems, second-order effects, incentives, psychology, and long-term consequences. That part is valuable. But I think people are missing something important: The moment you say “5D,” you’ve already admitted there could be 6D, 7D, or infinite dimensions beyond it. Because the dimensions themselves are not objectively defined. One person’s 5D could be: systems + time + psychology. Another person’s 7D could include: culture, evolution, civilization-scale behavior, recursive feedback loops, collective consciousness, AI-human coevolution. And suddenly the number becomes arbitrary. That’s why I don’t think intelligence is about collecting more “dimensions” forever. In fact, I think overdoing this creates a dangerous illusion: confusing complexity with wisdom. Some people become so obsessed with: hidden patterns, meta-systems, recursive abstractions, and “deep thinking” that they slowly detach from reality itself. Not everything is a hidden system. Not every outcome has a grand explanation. Sometimes randomness is the explanation. Ironically, the smartest people I’ve observed don’t endlessly expand dimensions. They compress them. They reduce complexity into a few leverage points that actually matter. That’s much harder. Anyone can make reality sound more complicated. Very few can simplify reality without becoming simplistic. Maybe real intelligence is not about reaching “higher dimensions” forever. Maybe it’s about knowing: which dimensions matter, which are noise, and when to stop thinking and start building.

@sarthakcore two weeks is about right. the launch doesn't teach you, the first real user complaint does. watching the docs.

formasty is live. ads are running. i'm documenting everything here. if you're building something and feel like you're too young, too early, or too inexperienced - you're probably 2 weeks from the thing that changes everything.

i'm 16. i built a SaaS with AI agents. not a tutorial app. not a hackathon project. a real product with paying users, MCP integration, and autonomous agents shipping code while i'm at school. here's how i did it: 🧵👇

human in the loop but mobile. the approval step is still a bottleneck, just a portable one. real shift is when intent is clear enough that nothing needs approving mid-flow.

@ComfySipper UI churn to hide a feature nobody asked for. moving the button doesn't fix that the button shouldn't exist.

Google maps has changed its UI like twice in a week to move where the useless and stupid AI button goes and I Hate It