Julien Codorniou

Announcing Slashwork.com, backed/incubated by @20vcFund and the very people who built Slack and Workplace. Proudly built in London. 🇬🇧

C'est comme ça (Remix by Tony Visconti & Les Rita Mitsouko) by Les Rita Mitsouko on #SoundCloud on.soundcloud.com/gwY2muCgVXkgYG…

@Carles_Reina slashwork.com exists to fix this, once for all.

The biggest challenge for any remote-first company is comms on Slack, where you are buried in messages, channels and pings. Specific messages are impossible to find, new joiners don't find the context after many threads, and everyone forgets it all. Add comms retention policies, and you have a ticking bomb. How are other companies fixing this?

19 years ago, a high school basketball coach put his team manager into a game for the final four minutes. The kid had never played a single minute of competitive basketball in his life. He scored 20 points. Jason McElwain was diagnosed with severe autism at age two. He didn’t speak until he was five. He couldn’t chew solid food until he was six. He wore a nappy for most of his early childhood. As a baby, he was rigid, wouldn’t make eye contact, and hid in corners away from other children. He tried out for his school basketball team every year and got cut every time. Too small. Too slight. Barely 5’6 and about 54 kilograms. But he loved the game so much that his mum called the school and asked if there was any way he could be involved. The coach created a team manager role for him. For three years, McElwain showed up to every practice and every game. He wore a shirt and tie on match days. He ran drills, handed out water, kept stats, and cheered every basket like he’d scored it himself. On 15 February 2006, the last home game of his final school year, the coach let him suit up in a proper jersey and sit on the bench. With four minutes left and a comfortable lead, the coach sent him in. His first shot missed. His second missed. Then something shifted. He hit a three-pointer. Then another. Then another. His teammates stopped shooting entirely and just kept passing him the ball. He hit six three-pointers and a two-pointer. 20 points in four minutes. The highest scorer in the game. When the final buzzer went, the entire crowd rushed the court and lifted him onto their shoulders. His mum tapped the coach on the shoulder, in tears. “This is the nicest gift you could have ever given my son.” McElwain won the ESPY Award for Best Moment in Sports that year, beating out some of the biggest names in professional sport. He’s 36 now. He works at a local supermarket, coaches basketball, has run 17 marathons including five Boston Marathons, and travels the country speaking about never giving up. When asked about that night, his coach still gets emotional. “For him to come in and seize the moment like he did was certainly more than I ever expected. I was an emotional wreck.”

Yann LeCun says today's humanoid robot demos look impressive, but the robots are very stupid in the real world "the missing piece is not hardware; it is AI that can't reason, plan, or adapt like humans" Robot companies are betting AI will make them smart enough to sell at scale within 3-5 years It's a big bet

Imagine you're John Carmack you're 22 years old and you just wrote a 3D engine in assembly that runs at 35fps on a 486 Doom drops. Quake drops. Half the planet is playing your code. you're the reason GPUs exist. you're the reason your friend Jensen has a yacht today. then in 2009, you sell id Software. people call it betrayal. you call it "they made an offer I couldn't refuse." VR obsession. Oculus. Meta buys it for $2B. you're CTO. but Meta thinks you're a liability. your demos are "too intense." your emails are "too long." your focus on frame timing is "slowing us down." 2022. they push you out. not fired officially. just "restructured." the media writes "end of an era." some crypto bro calls you "washed up." silicon valley moves on. but you don't. you don't write a book. you don't start a podcast. you don't collect speaking fees. you go completely quiet. you take the money. you buy a warehouse in Texas. you hire 10 engineers. and you start coding. not games. not VR. AGI. two years. radio silence. no tweets. no conference talks. while everyone's debating ChatGPT, you're debugging CUDA kernels at 3AM, testing world models. then in 2025, Keen Technologies pivots hard. you're not "exploring" anymore. you're building it. here's what people get wrong: everyone calls it a comeback. a redemption arc. "revenge on Meta." it's none of that. you're a 54-year-old engineer who still codes 12 hours a day because you genuinely can't stop. most CTOs would have bought an island. most legends would have written memoirs. you just kept typing. the most dangerous person in any codebase is the one who goes quiet and never stops shipping commits. karma doesn't need to be real. but obsession is. welcome back, Carmack.

How Founder and Employee-Led Branding Became the Best Channel for Lovable: "For any founder, I would suggest deploy founder-led social, employee-led social. Build in public, recruit people to be your biggest fans. And then you can start thinking about, okay, where else do I want to be? Do I wanna buy a billboard? Do I wanna invest into paid marketing? But first, organic strategy that every startup should be going into is employee-led." @ElenaVerna What does no one know that everyone should know about @antonosika @lukeharries @im_roy_lee @rauchg @karrisaarinen?

Lumineux.

@KillianMakaveli @DjibinhoPvris_ C'est mon frere

@DjibinhoPvris_ Je le considère comme un membre de ma famille

Ce que Paul Tchoukriel est entrain de faire pour nos vies wow

Yann just bet a billion dollars that the entire industry is building on the wrong foundation. Large language models predict the next word. They're trained on text, so they understand language. But the real world isn't made of words. It's made of continuous sensor data: camera feeds, touch, sound. And most of that data is unpredictable. You can't predict every pixel in a video the way you predict the next token in a sentence. Generative models fail here because they try to predict everything, including noise. AMI Labs is building world models using JEPA (a method LeCun proposed in 2022 that learns abstract representations of reality and predicts in that compressed space, not in raw pixels). Action-conditioned versions let AI simulate the consequences of actions before taking them. That's not generation. That's understanding. This unlocks AI that can operate in the physical world without hallucinating: 1. Robotics that plans multi-step actions 2. Healthcare devices where errors kill patients 3. Industrial process control under safety constraints 4. Wearables that adapt to real-time sensor input If JEPA works at scale, the next wave of AI companies won't fine-tune LLMs. They'll train world models on sensor data. LeCun's CEO already predicts every startup will rebrand as a "world model company" within six months. The architecture war is starting.

Aujourd’hui, 11 mars, mon cœur se souvient plus fort encore. Il y a 14 ans, mon fils Imad Ibn Ziaten, soldat de l’armée française, était assassiné par un terroriste à Toulouse. Il avait 30 ans. Il était un fils aimant, un frère, un ami fidèle, et un militaire fier de servir la France. Imad aimait la vie, il aimait son métier, et il croyait profondément aux valeurs de la République : le respect, la fraternité et la paix. On lui a arraché la vie, mais on ne pourra jamais effacer ce qu’il représentait. Depuis ce jour, je marche avec son souvenir dans mon cœur. J’ai choisi de transformer ma douleur en combat : un combat pour la paix, pour le dialogue, pour que nos enfants grandissent loin de la haine et de la violence. À travers l’Association Imad pour la Jeunesse et la Paix, je continue de transmettre son message et de faire vivre sa mémoire. Mon fils, tu es parti trop tôt, mais ton nom et tes valeurs continueront d’éclairer notre chemin. Imad, tu vis dans nos cœurs pour toujours. Que la France n’oublie jamais ses enfants tombés pour elle. association-imad.fr/faire-un-don/ #ImadIbnZiaten #Mémoire #11Mars #Paix #NeJamaisOublier #Maroc #France

Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia Khvicha Kvaratskhelia

After some time of using local AI cluster (Bob), here is my honest take on the good, the bad and overall use case. About a year ago I started playing with local AI models because of the work we do at BottleCap AI. I realised how amazing it actually is to own my own stack and my own data. At first, we used local models mainly because of security reasons as we do lots of AI efficiency research and new product concepts based on that. After OpenClaw was released, something changed for me. I started using local models much more, until they replaced cloud models for most of my deep-thinking tasks beyond work. Eventually, I canceled all my AI cloud subscriptions just to see if I could actually run fully on my local cluster. Hardware: • 2x Mac Studio with M3 Ultra and 512GB unified memory, 32-core CPU • 1x NVIDIA DGX Spark, added recently for prefills and, hopefully soon, faster inference • 10GB LAN Switch for connecting Spark and Mac Studio’s Current models: this is changing pretty frequently 1) “Bob OG”: • Main brain for reasoning and daily tasks • Qwen3.5-397B • Roughly 40-60 tokens/sec (depends on load & task) 2) “Bob Researcher”: • Long term researching • Qwen3.5-27B-Claude-4.6-Opus-Distilled-MLX-4bit: Very experimental 3) “Bob App Developer": • Coding apps and debuging • MiniMax M2.5 Software stack: • OpenClaw: All-local assistant layer • LM Studio: Running models • Exo Labs: Connecting multiple machines into one cluster and testing whether inference improves Where my local stack still lacks: • Deep tasks with big models still take more time to reply than cloud models. • Context window is limitation in the models I use. I’m usually around a 200k token window per session, but compacting works well, so I rarely need to start a new session. • It also seems that OpenClaw in its default state is not handling work with memory very efficiently while filling the context window fairly quickly by default. It was necessary for me to finetune this manually including semantic search and temporal decay which are in default switched off. • Reasoning is good but not at the cloud models level. Also coding is good for the majority of tasks but not top tier. My best use cases right now (March 2026): Best for iterative work where privacy matters and where model needs to be available all the time. • Private or sensitive data: I would be careful as a company to share private or direct customer information with third party cloud systems in general. Clearly also connecting OpenClaw to cloud models is not solving privacy situation. • Cloud limits & Efficiency: If I push cloud subscriptions hard, I hit consumer limits surprisingly fast. It’s also much easier to spot inefficiencies locally. When the context starts bloating, the system slows down fast, so issues like memory inefficiency become obvious much earlier. In the cloud, replies often feel just as fast, but you end up paying much more or hitting usage limits without really knowing why. Was it worth the money? For me, yes. But I’m aware I live in a niche bubble for my particular use case. For most people it is still early. For businesses and people who want to spend the money and effort make this work it is good solution today. My verdict: For my personal use case, local is now the default. Cloud is the exception. Are local models as good as the best cloud models? No. Are they good enough to be my default for most tasks? Yes.

Raising $ or investing $ is cool, but nothing, absolutely nothing, beats that feeling when you finally land your very first customer, fully deployed, super engaged, using & loving your product. The moment you realize you were not completely hallucinating, and that because it happened once, it WILL happen a million times. #slashwork

AI's Event Backbone altertable.ai/blog/2026-03-1…

I owe a lot of my career success to workplace, I recently found out that I’m the #10 most followed IC at Meta there. I’m proud to see that some of my ex-coworkers that built the original version are now making it their mission to help all the other companies around the world!

I first met @lxbrun almost 20 years ago 👨🏼‍🦳 at the Microsoft France HQ when he was building VirtuOz (his 1st AI company). Later we worked together for many years at Facebook after the acquisition of Wit.ai (his 2nd AI company). I was also fortunate to invest in the pre-seed round of @nabla_hq (his 3rd AI company). So when Alex started his 4th (!) AI company, teaming up with former Facebook colleagues/legends @laurentsolly, @ylecun, Michael Rabbat, Pascale Fung, Saining Xie backing them with @20vcFund was simply a no-brainer. @amilabs is building a new generation of AI systems designed to understand the world, reason, plan and operate with persistent memory, centered on world models. Global ambition on day one with teams in Paris 🇫🇷, Singapore 🇸🇬, Montreal 🇨🇦, and New York City 🇺🇸. We are super proud to be partnering with them.

I am joining @ylecun and an exceptional founding team to lead @amilabs as CEO. We have secured a $1.03 billion USD seed round to fuel our mission to build intelligent systems capable of truly understanding the real world—a long-term scientific endeavor.

𝗦𝘂𝗽𝗲𝗿 𝘀𝘆𝗺𝗽𝗮 𝗰̧𝗮 ! 😁 Pierre Sage offre au petit Maé, fan de Lens et plus précisément de Matthieu Udol, 𝘂𝗻𝗲 𝗹𝗲𝗰̧𝗼𝗻 𝘁𝗮𝗰𝘁𝗶𝗾𝘂𝗲 𝘀𝘂𝗿 𝗹𝗲 𝗿𝗼̂𝗹𝗲 𝗱𝘂 𝗽𝗶𝘀𝘁𝗼𝗻 𝗴𝗮𝘂𝗰𝗵𝗲. 😁📌⚽️ Super séquence ! 🎥 @ligue1plus