Hitesh Bhatt @B_HiTech
Self described Tesla-nut! State of Awesomeness Joined April 2016-
Tweets15K
-
Followers1K
-
Following745
-
Likes253K
Moroccan police are filmed forcing them into the water to cross into Ceuta. Making these most likely the criminals pardoned by their king last week, including terrorists!
BREAKING: Wall Street firms kept bullish ratings on SpaceX after its Q2 earnings call. Goldman raised its target to $220, JPMorgan to $240, and Raymond James kept its $800 target. Wall Street remains confident in SpaceX’s long-term Starlink, Starship and AI growth.
This is a big announcement and I’m not sure if people initially realized just how revolutionary this design and application is for @SpaceX and how these will be deployed both terrestrially and via Starship in orbit. It’s not just the fact that each of the Starmind satellites will include @nvidia Rubin GPUs and Vera CPUs for “Starmind” datacenter class space compute without having to wait for future joint venture @Tesla and @SpaceX AI chip development. It also signifies SpaceX and NVIDIA are able to build these Starmind satellites now in their existing facilities using the current V3 Starlink satellite format without having to wait for the AI Satellite Production facility construction in Bastrop, TX nor will they have to wait for the massive Terafab construction either. But, what is truly significant is these will be in a novel dual-use format that will be featured in both terrestrial and space-based AI Datacenters, a major breakthrough with packaging & reduction in costs and complexity. Essentially, this format will be identical for both applications with the space-based versions including solar cells and radiators. These will also feature the common “pancake” format that will work easily with the existing “Pez Dispenser” Starship payload deployment system used for Starlink V3 satellites, likely with little or no modifications! Truly Genius! @esherifftv @CernBasher @herbertong @thejefflutz @TeslaBoomerMama
SpaceX is partnering with @nvidia to design the Starmind AI1 satellite compute payload. Each of the Starmind satellites will include NVIDIA Rubin GPUs and Vera CPUs for datacenter class space compute → spacex.com/spacexai/starm…
@evgonetwork Congrats on making this shift to Tesla's v4 Superchargers for your network. I'm sure it'll get great utilization.
BREAKING: NASA chief Jared Isaacman says SpaceX’s fully reusable Starship could help build a Moon base that may one day grow into a small city. “We’re going to leverage reusable rockets to build the Moon base… You’ve got the heavy-lift vehicles, like what Starship is capable of doing, which is fully reusable… That’s what allows you to put lots of mass on the surface of the Moon. That’s what will take it from a small little outpost to potentially something that may even look like a small city someday.”
WOW … what a firehouse of a Financial Results meeting, the 1st for @SpaceX as a public company. So much to digest, bro start, let’s talk about Starship and how much of a big step up this will be compared to Falcon 9. First, the economics and capability offered when launching the V3 @Starlink Satellites as compared with the current V2 mini satellites launched by Falcon 9 are almost silly, Here are two slides I put together to help put this into context. For V3 Starlink launches, one Starship launch is about equivalent to 20-23 Falcon 9 launches! The more Starship matures, the better the efficiency becomes, and the time for payback is slashed by 75%! Cost per Kg of launch in about 10X lower overall, and this would extend to other launch payloads as well A few other datapoints from the Financial Results meeting: - They are now very confident in the Heat Shield solution they have in place and although more iteration work will happen, it is now good enough for reuse ability - Flight 14 should be by the end of August and will put the 1st V3 satellites into orbit and they will attempt to catch Ship 41 (with goal to catch both Booster and Ship before the end of 2026) - They have a goal of 1 launch per day in 2027 (distributed between Starbase and Kennedy Space Center) … this is likely a big stretch goal, but even getting to half of that (or even a 1/4 of that) would be almost incomprehensible in terms of mass to orbit, revenue generation with V3 Starlinks and time for payback! - They plan to meet their Artemis III commitment for LEO rendezvous next year and they stated a goal is to have Starship human rated by the end of 2027 😳 - They plan multiple landings on the moon in 2028 including returning humans to the surface for the Artemis program - The infrastructure they are putting in now to support Starship at both Kennedy Space Center and Starbase will be ready (enough with continued construction) to support the multiple Starship launches per week, if not daily. - Starmind satellites with NVIDIA GPUs will begin to be launched in 2027 (the orbiting AI Datacenter satellites) - Starlink Mobile satellites will begin to be launched in 2027 and will join with some kind of distributed Starlink dish system to provide cellular service all over (the start of the disruption to traditional cell service providers) Much more … I have to go through all of my notes and there are plenty more for Starlink, Starshield, AI & AI Compute services, government & enterprise opportunities for growth & more! What an exciting future to be looking towards … and it has already started! @esherifftv @herbertong @TeslaBoomerMama @thejefflutz
Here’s a @SpaceX Q2 2026 Earnings Call Summary if you missed it. It was a banger: • Elon Musk says it’s not out of the question that Starlink will deliver the majority of the world’s internet in less than 10 years. • Grok 4.6 is probably coming out next week, and Grok 4.7 is 3-4 weeks out from today. • Grok 5 will incorporate all of SpaceX’s data. Elon Musk says this will make Grok the best engineer. • SpaceX is deploying AI compute faster than anyone else. Their efficiency is also the highest. • SpaceX expects to end this year with 2GW of AI compute. • By the end of the next year, SpaceX’s cumulative AI compute will be closer to 10GW. • SpaceX has desired to build compute exclusively with Nvidia. AI1 satellites will use Nvidia Vera Rubin GPUs. • SpaceX expects to launch its AI1 Satellites in 2027. • SpaceX will deploy their satellite rack design on earth, and in space because it will be a more cost-effective design. Cool. • Elon Musk on the SpaceX team: "We have the smartest, most dedicated team of humans that’s ever existed." • SpaceX sees enterprise connectivity revenue substantially exceeding the consumer business. • SpaceX has never lost an enterprise customer before. • SpaceX is building an enterprise sales team to go out and meet with companies. • Elon says SpaceX and Tesla are the two best companies at hardware, including China. • Elon says SpaceX will receive a significant percentage of NVIDIA GPUs going forward. • SpaceX is expecting to reach $100B+ ARR by the end of this year. • Elon says their internal projections for SpaceX reaching $1T in revenue has been pushed up to 2030, not 2031. Bullish. • Elon says Starship 14 is scheduled for the end of this month. • SpaceX may catch Starship during Flight 14. • Elon said launch cadence is increased. A year from now he sees SpaceX launching Starship every day. • SpaceX predicts they will take acquire a lot of customers from AT&T, T-Mobile, and Verizon with their Starlink Mobile service. • Starlink Mobile service will be available by the end of 2027. • SpaceX would need about 1,000 Starlink V3 Satellites to begin seeing improvements in overall service. This could happen in Q2 2027. • Elon says Starship’s heat shield is pretty much a solved problem at this point.
@AbelAvellan @grok who launched these satellites?
@AbelAvellan Thank you SpaceX for the ride. 🙏
Full stack. Full circle. The Everything Company.
GOOD NEWS 🎨 Tesla has officially been granted a design patent (US D1,138,869 S) covering the distinctive exterior body of its Robovan concept. Filed back in October 2024 and officially issued on August 4, 2026, the filing credits Tesla Chief Designer @woodhaus2 and his team (Julien Bilodeau, Ian Kettle, and Joseph Robinson) with shaping the vehicle's unique look. A closer look at the patent illustrations reveals several striking design choices that shape Tesla’s futuristic mass-transit vision: 🚌 Mobile architectural volume: Rather than following conventional van proportions, the Robovan resembles an autonomous tram or a moving lounge. The exterior reads as a single, uninterrupted capsule optimized for passenger comfort. 🚂 Art Deco streamliner influence: With rounded corners, sweeping horizontal bands, and fully enclosed wheels, the aesthetic pulls heavily from Art Deco streamliner trains and buses—reinterpreted through a minimalist science-fiction lens. 🤖 Symmetrical, "faceless" profile: There is no conventional grille, hood, side mirrors, or driver cockpit. The distinction between the front and rear ends is so subtly muted that the vehicle visually communicates full autonomy at first glance. 🛸 Hovercraft silhouette: By concealing the wheels and undercarriage deep within the lower body panels, the vehicle visually hovers above the road rather than resting on four exposed tires. 🕶️ Wraparound glass envelope: A dark, continuous glass beltway wraps around the upper cabin. The curved termination points break up the sheer mass of the vehicle, making the high-capacity volume look surprisingly sleek. 💡 Horizontal stratification: Multiple parallel strakes run along the waist and lower sill. While patent line art doesn't specify materials, these bands are ideal locations for production lighting strips, signal indicators, or aerodynamic trim. 📦 Space-maximized packaging: Near-vertical front and rear end caps combined with full-width, full-height cabin walls maximize every square inch of usable interior space—making it equally suited for high-density ride-pooling or autonomous cargo logistics. ⚡ A distinct design language: Unlike the sharp, planar, stainless-steel geometry of the Cybertruck, the Robovan relies entirely on soft radii and continuous curvature, carving out a distinct "retro-futuristic public transit" aesthetic within Tesla's portfolio. Beyond the aesthetic novelty, this filing confirms that Tesla is actively locking down the intellectual property for its high-capacity autonomous fleet.
Looks ready to me. ‼️
🚨 SPACEXAI UPDATES FROM Q2 EARNINGS GROK: >Grok 4.5 was a huge improvement >Grok 4.6 dropping ~next week >Grok 4.7 in three or four weeks >Grok 5 before end of the year “We will be incorporating the ENTIRE CORPUS of SpaceX data… ALL the data SpaceX has ever produced over the course of a quarter century into Grok training… This will make Grok by far the BEST ENGINEER” COLOSSUS COMPUTE: >building data centers faster than anyone else >efficiency of compute deployment also the highest >2GW+ online by end of 2026 >by the end of next year… SEVERAL times higher >closer to 10GW than 5GW NVIDIA x SPACEX: >we’ve decided to build EXCLUSIVELY on NVIDIA >Vera Rubin architecture is the best architecture >“we greatly value a close cooperation and partnership on many levels with NVIDIA” CURSOR: >acquisition should close quite soon >through almost all of the regulatory hurdles STARMIND AI1 SATELLITES: >optimized Vera Rubin NVL72 in orbit >"we expect to start launching these next year" >a radical simplification of the normal NVL72 rack >will cost less, and be more effective >so deploying them on the ground AND in space SPACEX CFO: >robust demand for cloud services >increasingly favorable economics with each agreement we sign >supply-demand imbalance in the compute market will continue >less than ONE YEAR payback on new compute capex >Q3 already contracted $6.7B more cloud revenue >ramping starting in October “We believe this puts us on a trajectory, including contribution from Cursor, to reach 100 BILLION of ARR by the end of this year.” $SPCX
@TeslaBoomerMama Immediately! Currently have Verizon. My assumptions for Starlink Mobile: 1. Global connectivity standard 2. Cost of service is lower and simplified
AI compute is going to orbit. 🚀 @SpaceX’s Starmind AI1 satellite compute payload is powered by NVIDIA Vera Rubin NVL72, bringing AI factory compute closer to the stars. The next chapter of AI infrastructure boldly goes where no AI compute has gone before.
SpaceX is partnering with @nvidia to design the Starmind AI1 satellite compute payload. Each of the Starmind satellites will include NVIDIA Rubin GPUs and Vera CPUs for datacenter class space compute → spacex.com/spacexai/starm…
You might like
