À propos
As an enthusiast with co-founding experience I'm a builder of consumer-facing virtual…
Activité
4 k abonnés
-
Raphaël Theet a republié ceciProud moment for us at Reactor: partnering Day 0 with LTX.io on their newest realtime model Try it here: http://reactor.inc/ltxRaphaël Theet a republié ceciIntroducing LTX’s newest open-weight model for realtime AI avatars - available on Reactor day 0. We're proud to partner with LTX.io make it easier than ever to build lip-synced, fully-voiced AI personas at cinematic quality with our API. Try it here: reactor.inc/ltx
-
Raphaël Theet a partagé ceci+1Raphaël Theet a partagé ceciAfter 25 years building AI, computer vision, and 3D companies — three of them as co-founder, with exits to Autodesk and Unity — I'm open to my next chapter. I'm still the CTO who designs the hard parts himself and leads the teams that ship them. Most recently: generative-AI and computer-vision products as co-founder and CTO of AniML. What matters to me now is the problem and the team, more than the title. If you're building something ambitious in AI, let's talk.
-
Raphaël Theet a partagé ceciThis is how you bridge the gap between digital intelligence and physical reality. Pascal and Jensen are defining the new standard for Industrial AIRaphaël Theet a partagé ceciThis week in Houston, we crossed an important milestone for the future of Industry. We are proud to mark a new phase in the long-term strategic partnership between Dassault Systèmes and NVIDIA “Together, we’re bringing the Virtual Twin Factory with the AI Factory,” our CEO Pascal Daloz said on stage with Jensen Huang. This partnership is unique by design. We are not adding AI to Industry; we are building industrial AI on science. Dassault Systèmes brings science-based Virtual Twins, built on decades of industrial knowledge capitalized in models and representations. NVIDIA brings AI infrastructure and accelerated computing at global scale. This collaboration establishes a mission-critical industrial AI foundation. It turns science-validated AI into real industrial decisions at scale, across biology, materials, engineering, manufacturing and autonomous systems. 🔗 Press release https://lnkd.in/eF-aGf-a
-
Raphaël Theet a partagé ceciThe serious crazy guy building things in serious topics.Raphaël Theet a partagé ceciIt only needs a few crazy ones to fix a continent… Let's be crazy. We're launching PROTOTYPE: a new fund, fully focused on Europe. First check. First round. As early as it gets. With it we back what Europe is world-class at: robotics, automation, manufacturing, and anything that requires hard engineering. And we are crazy enough to be your first investor. Europe invented industry. We're the birthplace of precision manufacturing. The second largest manufacturing hub on the planet. Leaders in automation and robotics. And yet… We sell out our best tech to China. We export our best founders & most of our investment money to the US. That's insanity. What should we do instead? Build the next trillion euro companies in robotics, manufacturing, automation right here: in Europe. Showcase to young founders what is possible and change the system around them where needed. What we will do: → Publish all our fund updates and build in public: https://lnkd.in/d9n3Hujn → Showcase Europe’s Most Ambitious Startups on Youtube: https://lnkd.in/d584_n6Q → Launch & support more projects like EU–INC. Enough talking about Europe. Time to build it: Startups, makerspaces, student clubs, and much more. → And most importantly, invest into the best founders in Europe. Same model as our previous funds that are in their top 1-5% cohorts worldwide. We won’t do a VC fund in the classic sense. This will be a community of hyper-ambitious people who want to actively change Europe for the better. It only needs a few crazy ones to fix a continent… Let's be crazy. Check out www.prototypecap.com!
-
Raphaël Theet a partagé ceciUS tech doesn’t just offer solutions anymore. It buys the problems. And resets everything. Instead of trying to help aging manufacturers with software and processes, US tech just buys them out. As an operator-investor, merging Private Equity strategy with tech execution. That’s exactly what Amca is doing. Aerospace is booming globally, but it runs on a hyper-fragmented supply chain—slow, outdated, stuck in a pre-digital model. One of the last fortresses still operating as if time had stopped. At the end, they struggle to deliver components to giants like Airbus or Boeing. AMCA, founded by tech entrepreneurs, decided to say: Let’s change it all. No software patch. No step-by-step “transformation.” No endless committees. With $76.5M raised from top California VCs, they’re acquiring companies like Electro-Mech Components (founded in 1964 and still kicking, hats off!) And radically transforming them to deliver efficiently to Boeing and Lockheed Martin. Processes, software, execution culture, Everything goes full tech. Instantly. Think Tesla factories. (Whatever you think of the CEO, they’re way ahead.) No more “this is how we’ve always done it.” What’s new here is tech doesn’t just propose anymore, It takes over. Given the global tensions, this model of vertical integration between tech and industry might be the fastest, strongest way forward. Case in point: 400 French companies fight daily to meet the ramp-up for Rafale or keep pace with A320 orders. Google tried with Motorola. Microsoft with Nokia… Different era. Different scale. Here, it’s a different play: supplier by supplier. One by one. Bottom-up logic Got other recent examples in mind? #aerospace #entrepreneurship #HardTech #AdvancedManufacturing #Automation #FactoryOfTheFuture
-
Raphaël Theet a partagé ceciÀ chaque fois, cette mise en scène me frappe aux US. La moindre interview d'un CEO tech hardware se déroule devant un drapeau dans son usine. Des ateliers, des grosses machines, des fondateurs… et toujours ce drapeau omniprésent, surdimensionné, Incontournable. Et ce n’est pas un hasard : ils veulent qu’on le voie et ca marche! Astranis Space Technologies - Boom Supersonic - Anduril Industries - Hermeus - Hadrian - Rangeview (qq exemples de la cover) C'est présent dans les usines hyper high-tech de la Californie jusqu'aux usines de fusées du Texas. Peu importe l’État, le secteur ou le bord politique. Les Américains racontent leur industrie comme une épopée, comme ils l'ont toujours fait. C'est culturel, ancrée dans chaque entreprise, qu'elle soit jeune ou ancienne, high-tech ou industrie lourde. C'est une stratégie naturelle et organique trans-partisane qu'ils possèdent de façon innée, c'est simplement américain. Une sorte de propagande naturelle un prolongement d'Hollywood appliqué à l'industrie. Ce drapeau symbolise un renouveau manufacturier aux États-Unis. Le « New Industrialism » américain se reconstruit par le récit commun. Et l'Europe ? Oui, l'appartenance au drapeau européen reste difficile (bureaucratique et abstrait) et l'imaginaire ne suit pas encore. L'Europe possède les grands programmes, les ingénieurs, les usines, mais pas encore de narration collective. La mise en scène, la posture et le récit stratégique restent encore à inventer au-delà d'Airbus. Ce n'est pas nouveau, la prudence de l'Europe envers les symboles l'a privée de la capacité à dire « nous » et à créer une stratégie narative . Peut-être que le monde qui change aujourd’hui changera cela. (Shorter English version) Every hardware tech CEO interview in the US features a flag prominently displayed in their factory. Workshops, machines, founders... and always that flag. Omnipresent and unavoidable. And it’s no accident: they want you to see it and it works! This happens in both California's tech factories and Texas's rocket facilities - regardless of politics or industry. Americans portray their industry as an epic journey. This approach is cultural, integrated into every company's DNA. It's a natural cross-partisan strategy they possess innately - simply American. Like Hollywood applied to manufacturing. This flag represents America's manufacturing revival through shared narrative. What about Europe? Connection to the European flag remains difficult (bureaucratic and abstract). Europe has the programs and talent but lacks collective narrative beyond Airbus. Europe's caution toward symbols has limited its ability to create a unified story. Perhaps today's changing world will change this. Examples of New Industrialism CEOs : AJ Piplica - John Gedmark - Blake Scholl - Palmer Luckey - Chris Power - Cameron Schiller #engineering #Manufacturing #industry #ingénierie #MadeInEurope #MadeInAmerica #industrie
-
-
Raphaël Theet a partagé ceciSimple products take time, a lot of time and a lot of energy. For some teams, simplicity becomes a strategic obsession and the value unlocked is huge! Spoiler: The smallest engine at the left (A) is also… the most powerful, the least bulky, the least fragile, the cheapest, the simplest, and the fastest to build. Reducing a product down to its bare essentials while increasing perceived value is insanely hard. Just look Enzo Avigo who constantly simplifies his SaaS product June always anticipating complexity before it even sets in while adding value every week. Simplicity → Complexity = Natural Any product naturally gets more complex. Not the other way around. Most SW roadmaps go from super simple (feature company) to super complex (platform). It's by far the standard path for most tech success stories today. Eventually once it's too complex the CEO sells to a bigger fish, a company built to deal with this level of complexity and to sell to clients just as complex. The Full "Enterprise private cloud suite" SW market. That's B2B SW life. It works. It's fine. But some engineering teams, that don't have choices, get so obsessed with simplification that they manage to shrink the size, the weight and the complexity while boosting performance. An amazing product work! This engine wasn't simplified for fun: fewer parts mean less bulk, more engines on the rocket, fewer failures during production, 4x lower cost and a production rate of 1 per day. That's impressive. Even though it's smaller and lighter the version 3 of the Raptor ( "A" the tiny one on the left in the pic) is actually the most powerful. "Raptor 1" 180 tons of thrust. "Raptor 3" 280 tons! The craziest part? These are the same type of teams who built the Raptor… and blew up launchpads, failed stage separations, upper stage breakups. Yes this kind of simplicity takes huge amounts of time and money (thanks to the US Air Force and American public funding...). Sure It's expensive, it's not for everyone but it's worth a look simplicity is so hard. #engineering #productdesign #mechanicalengineering #aerospaceengineering #spacex Source : SpaceX NB : This article focuses on engineering accomplishments, not politics. SpaceX employs brilliant engineers who deserve credit for these innovations. This post acknowledges the technical achievements, not any political positions.
-
Raphaël Theet a partagé ceciIt's getting closer and closer to human workflow. The Figure 02 humanoid robot performed 60 minutes of sorting without any human intervention. It detects, flips, and aligns packages, even the soft ones, to optimize barcode reading. It doesn't follow a pre-programmed script: it observes, analyzes, makes decisions, and adapts in real-time, exactly like a human operator. The envisioned business model is : customers pay an hourly fee covering all operational costs (maintenance, electricity, insurance...) making it easy to compare with human labor costs. It gives food for thought. #Automation #Robotics #HumanoidRobots #ArtificialIntelligence
-
Raphaël Theet a aimé ceciRaphaël Theet a aimé ceciThe question I get most in sales calls: Twin or Claude Cowork? Honest answer: they do different jobs. Cowork is built to work alongside you. It’s good when one person is doing focused, hands-on work. Twin is built for the work that should run without you. The recurring jobs across your apps, on a schedule, in your Slack. A few things that shaped how we built Twin: 1️⃣ Approach Cowork can run tasks on schedule, but mainly builds one off agents every time. 🔸 Twin fires AI employees that own a process end to end until the job is done. 2️⃣ Access to tools Cowork picks from a connector catalog. 🔸 Twin builds and maintains its own tools. It connects itself to 6,000+ apps and everything web, it builds scrapers and APIs. No MCPs needed. 3️⃣ Models Cowork runs on Claude models. 🔸 Twin picks the best and cheapest model for each task across multiple providers. Costs drop up to 90% because it uses open source models on scheduled tasks. 4️⃣ Where it lives Cowork sits in the Claude app. 🔸 Twin lives in your Slack, Teams, WhatsApp and Gmail, so anyone on the team can ping it from anywhere. 5️⃣ Building & sharing With Cowork everyone tends to rebuild their own thing from scratch. 🔸 On Twin your team can pick from our org template, reuse them and we can also auto-route anyone to an existing agent to reduce costs of duplication. So my honest opinion: Pick Cowork if you want a thinking partner or copilot. Pick Twin if you want to delegate work to autonomous agents. They just solve different problems. PS: send me a DM and I'll get Twin FREE credits for you 💛
-
Raphaël Theet a aimé ceciVery excited for this release: drag to edit in realtime, on video. Only on ReactorRaphaël Theet a aimé ceciX2 is now live on Reactor, the newest real-time video model from XMAX. Drag directly on the video to shape motion as it plays. Instantly swap characters, outfits, or art styles while perfectly preserving the original motion and lighting.
-
Raphaël Theet a réagi à ceciWriting angel checks to whomever is here on a Sunday at 7am. DM me
-
Raphaël Theet a aimé ceciRaphaël Theet a aimé ceciEnjoying this brief moment in time when the book is #1 best seller in France across all its categories 😍 🙏
-
Raphaël Theet a réagi à ceciRaphaël Theet a réagi à ceciI know this will sound crazy, but this IBM ad from 1951 says a lot about our future with AI. Look at the headline: "150 Extra Engineers." IBM was selling an electronic calculator and arguing that one machine could do the work of 150 additional engineers. At the time, engineers performed many calculations using slide rules, the analog computing devices shown in the ad. IBM's machine automated a large part of that work. It occupied an entire room, cost a fortune, and could perform operations that seem trivial by today's standards: addition, subtraction, multiplication, and division. In a way, it was the Claude Code of the early 1950s. Less capable, far more expensive, and with much lower distribution, but built around the same promise: automate work that previously required skilled professionals. What happened next? Did IBM eliminate engineers? Quite the opposite. The demand for engineers exploded in the decades that followed. Some tasks disappeared, some jobs changed, and entirely new specialties emerged. The calculator did not reduce the need for engineering talent. It increased what engineers could accomplish. That does not mean AI will be painless. Some jobs will disappear. Some people will be forced to adapt. Every major technological shift creates winners and losers. But when I look at history, I find it hard to believe that increasing human productivity leads to less opportunity overall. My bet is that AI will create far more value than it destroys. The real question is not whether work will change. It's whether we will adapt quickly enough to take advantage of the change.
-
Raphaël Theet a aimé ceciRaphaël Theet a aimé ceciHoly crap! Apple just beat Google to the punch -- 3D Gaussian splatting is coming to Apple Maps. These 3D scenes are made from oblique aerial imagery. But unlike blobby photogrammetry -- no more broccoli trees, no more melted powerlines -- ground level detail that actually holds up. Here's hoping Google Maps/Earth follows suit soon -- they have a significantly larger corpus of sensor data to work with. Time to splat the world!
-
Raphaël Theet a aimé ceciRaphaël Theet a aimé ceci"You click, and it turns" 😅 SKETCHFAB - A Startup Memoir 📘 Available soon 👀 "Around the same time, we entered another competition, this one closer to home: “Le Cercle des Jeunes Entrepreneurs,” a gathering of young founders in Paris. We won. The prize was a check for fifteen hundred euros. It was, technically, our first external funding. I deposited it with the gravity of someone handling a much larger sum. The competition also required us to man a booth for the evening, which was a new experience. I found it energizing - the chance to demo the product live, to watch people’s faces when a 3D model materialized in their browser. Cédric found it considerably less energizing. He stood behind the table with the resigned patience of a man waiting for a delayed train, answering questions when asked but volunteering nothing. As the evening wore on and the crowd thinned, one last visitor wandered over. He looked at the laptop screen, then at Cédric, and asked what this was all about. Cédric had been standing at that booth for five hours. He was done. He looked at the man, looked at the screen, and delivered what might be the most concise product pitch in startup history. “You click, and it turns.” The man nodded slowly, as if this explained everything, and walked away. “You click, and it turns” became part of the Sketchfab vocabulary from that night forward - our unofficial tagline, an inside joke, and, if you stripped away everything else, a perfectly accurate description of what we’d built. Sometimes the best pitch is the honest one. Sometimes, when you’re exhausted and you’ve stopped trying to impress anyone, you accidentally say the truest thing. We had fifteen hundred euros in the bank, a growing community of users, a standing invitation from one of Europe’s most legendary investors, and a product that, when you clicked it, turned. It was time to get serious." #startup #book
Expérience et formation
Langues
-
French
-
-
English
-
Recommandations reçues
5 personnes ont recommandé Raphaël
Inscrivez-vous pour y accéder