I’m watching an NVIDIA GTC from 9 years ago. An IISc based startup presented a GPU native CFD solver at the conference. Today ANSYS Siemens Cadence are all jumping in to deliver GPU based CFD Solvers. The Indian company was way ahead then and still isn’t a name without even 1% market share and maybe just 5-8% mkt share in India. What failed them?

The Dispatch Chronicles We made more weapons than anybody in India. Now we had to deliver several 1000’s of them & more than 3 variants on the same day Zero Hour minus 72 hours | 09: 00: "We'll wrap this up by tomorrow night." (Famous last words.) Zero Hour minus 40 hours: The chai vendor is the most loved guy in the plant. He apparently knows about all weapons & is planning on writing a book on us. Good bloke, he makes it a point to to ask “Aaraama?” & reduce my BP Zero hour minus 30 hours: Someone tried to scan a samosa for a serial number. Zero hour minus 24 hours: The packing list & test reports are now longer than most novels. We've read it more times than any novel. Zero hour minus 14 hours: We’ve decided to promote the forklift to Senior Team Member. It's the only one still running at full capacity. Zero hour minus 7 hours: Break time. Our 2IC of testing decides it’s time to go ballistic. We don’t protest. He’s a Special Forces guy & the sweetest among us Zero hour minus 3 hours: We’ve not seen so many different weapons ready on the same dispatch day. Every crate sealed, every serial matched, every document signed. Zero hour plus 15 hours: Just heard there’s someone sleeping in the conference room. Our resident canine unit found him & sent him home.

Up next, Falcon Heavy will launch the NROL-97 mission from Launch Complex 39A in Florida → spacex.com/launches/nrol97

Today, at Secretary Hegseth's direction, the Department of War is starting the creation of the Autonomous Warfare Command. This will be a new 4-star combatant command—with service-like authorities—built to scale autonomous and robotic capabilities across the Joint Force in the fastest peacetime shift in modern military history.

Map of all active & planned Solar System missions and their destinations as of October 1st 2026. Happy new year Martians! @mmx_jaxa_en launch to Mars scheduled on the 10th #keepExploring

My 2 cents on the hype of rockets. The small satellite launch market, is very small. Let's say we are to launch a 100 satellites, 50kg each @ 7000$/kg , we end up at 35M$ revenue. 35M $ (350cr INR) is the actual addressable market. Now that is indeed really small. In this market, there are more than 25 companies across the world building rockets. It is innocent of us to think that the global market is really global and shall stay global. Today there are no widely succesful options. But upcoming... => Japan alone has Space One, Honda and Interstellar Technologies. => India has Skyroot Aerospace and Agnikul Cosmos. => Germany has Isar Aerospace, Rocket Factory Augsburg and HyImpulse. => France has MaiaSpace, Latitude, Sirius Space Services, HyPrSpace and others. => The UK has Skyrora and Orbex. => Spain has PLD Space. => South Korea has Innospace and Perigee Aerospace. => Australia has Gilmour Space. => Canada has Reaction Dynamics and C6. => The United States has Rocket Lab, Firefly, ABL Space Systems, Astra, Phantom Space and several others. => And then there is China — Galactic Energy, i-Space, LandSpace, Space Pioneer, Orienspace, CAS Space and others. When all the countries eventually succeed in next 3-5 years time frame, the globalisation will slowly cease to exist. Why will country A allow its satellites to be launched via country B vehicle? A will try to give business to it's own industry rockets. Given such hot space & Incredible competition , the 35M market, even if there is demand for 1000 satellites, the entire market will end up at 350M and still wouldn't be relevant enough revenue for any single company. Now this is the scenario after success across geographies. But today the scenario is different. Today small launch is widely done via RKLAB Electron. It did 190M$ revenue in launch segment across 21 missions. Which indicates for a PL Capacity of 300kg, the avg cost was around 28lakh/kg. This shows that, for individual sats and orbit control, price/kg is a premium and $/kg is rarely a metric that is tracked for small rockets. But then, RKLAB's filings show, it was able to bring down the cost per launch to 4.8M$ , while average revenue per launch is 8.5M$, which is 44% gross margin in launch. With other options getting widely available, the gross margin may go down to 10% at max. Which means, in today's market conditions, 21 launches @ 5.3M(after margin 10%) ~= 110M$. With a CAGR of 10% , this after 10 years is 260M$. Let's even take that market after 10 years 350M$. Almost all the companies in above list are currently valued at > 4X this small satellite launch market 😂 So at this moment, there are 40+ companies fighting for this market with combined valuation well above 50x the market size that will be 10 years later 🤣 Definitely, there has to be something more to it right? Sure. Every rocket company has plans of bigger rockets. Reusable rockets. Bigger engine rockets. Different rockets. This is the promise they show investors. We shall become Falcon9. Nice & Interesting. But end of day, Falcon 9 has become Falcon 9 because of US & Starlink. Incredible high tonnage US demand combined with profitable Starlink created the recurring demand for heavy weight vehicle. Will each of the above companies also field a mega constellation? May be every country will field its own mega constellation, right? Sure, so the entire premise of extremely bulky valuations rests on the fact that Governments/Entities shall create demand for 1000s of Heavy satellites in each country. The rocket companies can eat the demand. Then all is well for every company. This i don't think is the scenario in next 3 years atleast. May be 10 years later? Sure. So till then, these are strategic bets forming a big bubble. Elon musk has starlink & XAI to shield from smaller launch business. Launch gives the "Frontierness" to SpaceX brand. RKLAB has 3 times bigger space systems business over launch revenue. Same case like SpaceX Amazon has the mega constellation for its own vehicle. People who don't see the reality, that launch is small, will eventually fiddle out. What drives your launch is more important than the launch itself. Globalisation is a myth. Space is not an FMCG market to make in country A and sell in country B. It is an extremely important strategic market, and once internal capability is developed, rarely will demand fly to other countries. Countries will protect their industry, once they succeed. Small satellite launch market is very small, and eventual ride share will eat small rocket market. Big rockets, Mega constellation, Space infra is the future. The future of launch belongs to whoever controls recurring payload demand, not necessarily whoever builds the smallest or cheapest rocket.

I don’t think people realize how BIG SpaceX @Starlink V3 Satellites are when deployed. When stowed inside Starship, they’re 7 m wide. When deployed in Orbit, Starlink V3 Satellites have a 60 m (197 ft) wingspan, which is the same size as a Boeing 787’s wingspan. Huge. Starship is designed to carry SIXTY of these.

So, @etherealmachine has made its own 5-axis CnC machines from scratch in India. Quite remarkable, really. We got the exclusive look at their factory

Maximum City Mumbai isn’t a place. It’s a pulse.

🇮🇳🚀 India's Indigenous Micro/Small Jet-Engine Ecosystem — The Full Map WHAT: A rare, comprehensive look at how deep India's indigenous micro and small jet-engine base has actually become — spanning government labs, PSUs, and a genuinely broad private industry base, all building propulsion for drones, loitering munitions, and cruise missiles. WHY THIS MATTERS: We've covered pieces of this ecosystem individually — DRDO-GTRE's Kaveri Derivative Engine, CSIR-NAL's NJ series, Prime Toolings' RDE work — but seeing them mapped together shows something bigger: India isn't just building one hero engine, it's building an entire propulsion supply chain across multiple thrust classes simultaneously. THE FULL PICTURE: 🔹 Government/PSU labs: CSIR-NAL (NJ-5/05/50/100 family, 5-100kg thrust), DRDO-ADE (drone/missile turbojets, 1-4.4kN class), DRDO-GTRE (350kgf ATGG engine + Kaveri/AMCA work), HAL-AERDC (HTFE-25, 25kN turbofan), DRDO-RCI+NAL (Rustom-II engine, ~2.7kN) 🔹 Private industry — genuinely thriving: Azad Engineering (already delivered a 350kgf engine to GTRE, July 2026), Raghu Vamsi's INDRA RV series (14-40kgf, flight-ready), DG Propulsion (20-60kgf compact turbojets for UAV/loitering munitions), Poeir Jets, Bharat Forge, Nabhdrishti Aerospace (ND400), DheyaTech (20-400kgf range, 2026 flight trials planned), Hycosys (hydrogen micro gas turbine, longer-term) WHY IT MATTERS: This is exactly the kind of depth that sustains a real wartime production base — not just headline platforms, but a genuine multi-vendor ecosystem across thrust classes (5kgf micro-turbines up to 350kgf turbojets), reducing single-point-of-failure risk if any one programme stalls. Combined with what we've tracked — Prime Toolings' RDE work, KDE's near-certification status, Aravalli's JV model — India's propulsion base is shaping up to be broad enough to actually sustain sortie-generation and munitions production at scale during a prolonged conflict, not just produce impressive prototypes. #IndigenousEngines #DRDO #GTRE #CSIRNAL #AatmanirbharBharat #MakeInIndia #IndianDefence #DefenceWithIndia