Vibhor Agarwal

Entirely right → India does not have a talent problem in deep tech. For decades, the global narrative was that Indian engineers were only meant for enterprise software, databases, or low-margin contract work. We were told advanced metallurgy and cryogenic infrastructure belonged strictly to Western or Chinese state-backed monoliths. Astrobase is smashing that myth on the concrete floor of their own test facility. We never lacked the minds. We have the toolmakers, fluid dynamics geniuses, and systems engineers to stand toe-to-toe with anyone in Munich or California. What we lacked was the institutional permission to compete, venture scale, and sheer audacity to look at a thermodynamic suicide mission and say → "We will build the factory, we will lay the concrete for the test stands, and we will fire it ourselves." The stands are ready. The turbopumps are clearing flow tests. The Indian factory floor isn’t just assembling the future anymore; we are generating the thrust to launch it.

@kamayanitweets 🤣

Bharat is not the future, it is here!

@SarvamAI @hcltech @BessemerVP @khoslaventures @peakxvpartners Way to go!

We're thrilled to announce that we have raised $234M in the first close of our $300M Series B at a $1.5B valuation. @HCLTech and @BessemerVP have joined us in this round, alongside continued support from @khoslaventures and @peakxvpartners For countries and companies, sovereign control on the AI stack is no longer an optionality. Sarvam will be the partner of choice for this aspiration. The capital allows us to accelerate our momentum towards this full stack of models, compute, and deployments. A huge thank you to our customers, partners, investors, and the Sarvam team for your trust and belief in what we are building. We’re just getting started. Read more: sarvam.ai/announcing-ser…

Moin's call-out to the ecosystem highlights our biggest national macro-shift. For decades, private Indian players were treated like peripheral suppliers - trusted with basic brackets but never the core architecture. Privatization didn't just clear red tape; it unlocked global credibility. When an IIT Madras incubator startup pitches custom, reusable launch configurations directly to European buyers, the old narrative dies. India is no longer just a destination for low-margin contract assembly. We are exporters of deep-tech intellectual property. The space race used to require a superpower’s treasury. Today, a handful of engineers from Chennai are proving that with pure shop-floor grit, you can conquer orbit on your own terms.

When you want to build a world-class, multi-cavity hot-runner tool that molds hyper-precision engineering components for global supply chains, you do not buy that machinery on day 1 of your business. If you try to finance a multi-crore imported press when your working capital is zero and your plant floor is a dusty shed, the bank will liquidate you before you can even hook up the main power. You start by running simple, high margin, high demand utility items. You mold unsexy plastic containers, basic industrial crates, or simple bottle caps. You let the machine run non-stop, 24/7, spitting out steady, predictable cash flow. You use that unglamorous volume to level the concrete floor, stabilize the erratic power grid, train your operators to respect microns, and build an unshakeable capital reserve. Only then do you use that financial muscle to fund the high-risk, zero-tolerance deep-tech moonshot. Aviral's brilliant essay on the reality of Indian ambition hits like a hammer on a cold anvil. It exposes the utter delusion of critics who don't understand the basic physics of capital sequencing. Pressure of Wealth Internet has democratized access to raw ambition, but it cannot magically alter the macroeconomic laws of thermodynamics. Look at the stark, uncompromising numbers highlighted in the piece → The US Ecosystem → A $25 trillion economy backed by $160 trillion in systemic wealth. The Indian Ecosystem → A $4 trillion economy with $16 trillion in baseline wealth. To the US, incinerating a billion dollars on a failed AI model or an exploding prototype rocket is a rounding error. It’s just minor friction. But to a $4 trillion economy, burning that same billion dollars is a painful, deep structural wound that bites directly into national survival. Our entrepreneurs cannot afford to play the lottery with unhedged capital; they have to align their risk perfectly with the actual market maturity of the country. "Chooran" Catalyst for Moonshots Without the massive consumer engines that solved immediate local needs, there would be no sovereign venture capital left to fuel our aerospace labs, electric propulsion engineers, or domestic defense startups. You need the high-volume consumer cycle to generate the financial pressure required to mold the deep-tech future. Industrial Evolutionary Curve History is an uncompromising teacher of sequence. No country ever leapt directly from an agrarian baseline to an aerospace superpower overnight. They all crawled through the mud of unsexy, low-margin manufacturing first. Ambition is entirely contextual. Solving immediate, ground-level structural needs for a population emerging from the basics is a massive, high-stakes achievement. A $1 billion Indian aerospace startup emerging from a resource constrained environment takes just as much grit, if not more, than a $100 billion defense giant backed by the endless deep pockets of a superpower. Ultimate Shop-Floor Comeback When an elite investor sitting in a climate controlled corporate tower told Sachin Bansal that he wasn't doing anything truly meaningful because he was just adapting a Western model for India, Sachin didn’t launch into a defensive pitch. He looked him in the eye and delivered the ultimate shop-floor response → "Haan, woh sab theek hai. Tu karke dikha." (Yeah, that's all fine. You do it and show me.) It is the definitive war cry of the Indian entrepreneur. It is easy to draft pristine, theoretical blueprints on a whiteboard. It is easy to lecture from the sidelines about sovereign AI and rockets. But until you have stood on the concrete floor, faced the raw friction of localized regulations, managed the cash-flow cycles of a developing market, and kept the assembly lines moving against all odds, you don't know the reality of the hustle. We are moving up the value chain at our own pace, using our own physics, on our own terms. Until the critics can step onto the shop floor and pull the lever themselves, the builders will keep running the machines. For a deeper look into how these unglamorous consumer tech gains are directly funding next-generation physical engineering in India, check out update on Deepinder Goyal's LAT Aerospace Venture. This video explains how capital generated from consumer platforms is now actively pivoting into building regional air mobility and advanced propulsion systems within the country.

James Dyson on the most significant lessons of his youth:

100% learnt it first day when I put my hands on communication modules

Think about a sourdough starter sitting in a jar on a bakery counter. It’s not just flour and water; it’s a living history of that specific kitchen. It captures the wild yeast floating in the air, temperature of the room, and literal touch of the baker’s hands. Every single day you bake, you take a little out, feed it a little more, and the flavor deepens. You can’t buy that exact taste at a supermarket, because it only exists as a continuous, compounding loop between the baker and the jar. Right now, most companies are treating AI like a fancy microwave they bought from Silicon Valley. You plug it into the wall, press a button, and it spits out a finished product. It’s fast, it’s clean, and it feels like magic. But if every company in the world is plugging into the exact same three microwaves, nobody actually learns how to cook anymore. What’s being argued here is the difference between renting a machine and keeping your own starter. The post uses corporate terms like "human capital" and "token capital," but the names barely matter. Human capital is the baker → taste buds, intuition, memory of how a bad batch felt, relationships built over decades. Token capital is the jar → AI system that absorbs those specific habits & keeps them alive. The danger isn't that AI will fail; it's that it will get so good at the generic stuff that companies will willingly hand over their own recipes. It is exactly what happened with globalization a few decades ago. On paper, shipping the factory overseas made perfect sense. The spreadsheets looked beautiful. But when the factory left, the town didn't just lose jobs; it lost the unwritten knowledge of how to weld a specific seam or fix a temperamental machine. The core capability was hollowed out, leaving a ghost town behind. If you just rent a frontier model to do your thinking, you are outsourcing your company's brain. The moment they change the algorithm or raise the price, you realize you own nothing but a monthly subscription. The real play is building a "hill climbing machine." You want an AI system that doesn't just generate text, but actually watches your company's veterans work. It needs to capture the weird shorthand, the hard-learned lessons, and the private fixes that never make it into the official training manuals. You bake that unique judgment into your own private loop. If a better general model comes along next year, you just swap the underlying engine out like changing the brand of flour you buy. The starter in the jar stays yours. So picture your company not as a sterile office full of people using digital tools, but as an ecosystem where every single task fed into the machine leaves behind a tiny grain of institutional memory. The goal isn’t to let a few massive tech giants own the global kitchen; it’s to make sure your own jar keeps fermenting, getting sharper, more specific, and utterly impossible for anyone else to copy.

Refreshing weekend, back to grind

@RaunakPahwa Wish this was implemented before my XR was stolen 🥲🥲

@vibs98 Rumors are even parts will be blocked and in whatever device the parts are used, then that phone will get blocked😭

If this happens, I will switch to Apple ecosystem

Nothing is really expensive when it comes to health; given it adds value

@FIR31415 @VecrosTech @besta_tweets 👀👀

@FIR31415 @VecrosTech @besta_tweets Launch it inside a fully sealed shipping container that is mounted on a moving truck

we built a drone that flies without GPS, without a pilot, without any signal tested it in darkness, in GPS-denied zones. it just… works. so here's an open challenge: name an environment you think it can't handle. we'll test the best suggestion on camera.

@siddharth_ram @PramaanaLabs @vkhosla Love this

We premiered the launch video of @PramaanaLabs at the Verification Summit in presence of @vkhosla. Many folks could not believe that it was written and made by us from India. Kept guessing names of agencies in US who could have made it. Not just technical talent but creative talent too is truly going global from India in this AI era. It’s time for Global Indian Ambition. Crafted in India and Premiered in SF

@SagarAwatade How do I preorder this? I have been looking! 👀

Our ancestors cleaned their teeth with a Neem stick. Then threw it in the garden. It became soil. We clean our teeth with plastic. We throw it away after 3 months. And it lives for 500 years. That's why we're experimenting with Neem, Miswak and Babul. The goal is to create a toothbrush that is ecological, affordable & sustainable. Should we bring back ancient oral care traditions and reinvent them for today?

Starting a company Scaling a company

@oprydai Forklift & a robot

man only want two things

@koushikBusim @oprydai Yes

@oprydai Warehouse & robotics? Or Women & robotics?😂