Mark Chitty (@TechPdM) header
Mark Chitty (@TechPdM) profile picture

Senior Product Manager @Zap_Map. In digital since the late 90s. Fascinated by #EV,#AI,#AR, #VR. Agentic dev.

Bristol, England
Joined June 2019
Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)Photo by Mark Chitty (@TechPdM)
This guy's animations which address the vertigo of agentic development you feel as a developer are absolute fire. youtu.be/a1e2ZDW3pKo?is=rizJ…
4
Mark Chitty retweeted
I'm in my autonomous car, which is driving me to Robot Fights in San Francisco. I've seen enough startups now and enough papers to know that wondrous things are about to come. But we did open Pandora's box to a whole range of new human problems. For instance, misinformation: it is far more powerful in this new world that's about to come than it is in the current world. That's not me saying that; that's Jeremy Bailenson, who runs VR and AR at Stanford University. And that's just one of the issues that we're going to have to face as humans. It leads to brain-computer interfaces, which you'll start using once you get into the glasses. You all will, because it's a far superior way to compute than looking at your phone, even if your idea of computing is watching YouTube videos all day long. Lately, I've been meeting people who are like, "I'm using glasses to run my agents." Viture glasses give you a 50-inch screen, or maybe even a little bigger, in front of you. So imagine being in an autonomous car and having the entire visual field in front of you augmented. In Silicon Valley, we have a millimeter-wave radar sensor company. I did a video with them two years ago at the Consumer Electronics Show. In smoke, fog, or driving snow, that sensor can see the road ahead for about 300 yards in 3D, even if the human eye can barely see the road. It costs $400 plus plus plus; let's say it turns into $1,600 once it gets to the retail level. But if Elon figures out he should have one of these in the Cybercabs in places where there could be snow, fog, smoke, or dust, he can make his Cybercab/robotaxi network far more reliable. That will make him more money and easily cover the $1,600 cost (or $5,000 if you include the software). We're also about to get characters that can talk to us (you've seen me share quite a few of those). That'll be a big deal in the glasses, right next to your agents. That leads directly to robotics, where you can put that digital twin on top of your robot. Now you've just given your virtual being eyes, ears, touch, and pretty much everything else, because anything it can see, it can figure out. If Elon's team gets this all working, Optimus is going to be a bigger deal than I'm expecting. But that's an aside. Wondrous things are about to come. Let's figure out how to keep from destroying humanity as well, please.
Humans have created lower dimensional models to reason and communicate about the natural universe: language, math and digital compute. Computers are on their way to being much better than us for all these domains. And at a time when scientific discovery appears to have slowed down relative to investment. We needed this new tool to continue to understand and control natural reality. It's not clear how far it will get us. But certainly another great age of discovery is here.
13
5
2
64
7,654
Mark Chitty retweeted
Things I learned after 4 kids and lots of shitty baby products and advice, and what I'd do now if I started again from kid #1: - Find a pediatrician you can text. Best money you'll spend in year 1 - NO crib. Get the biggest mattress you can fit on the floor, nobody falls off a floor. Newborns still get their own safe sleep space right next to you - Night shifts. One parent takes the night, the other sleeps in another room with earplugs, then swap. Two zombie parents make terrible decisions - Blackout blinds + white noise machine in EVERY kids room. If the sun wakes them at 5:30, the whole house is up at 5:30 - Zipper pajamas, NEVER snaps. Nobody wants to do 14 snaps at 3am with one eye open - Don't buy baby clothes. You'll get like 200 pieces as gifts and they outgrow half of them with the tags still on - A good baby carrier beats any stroller. Coffee shops, museums, beaches, airports, stairs, cobblestones. Try it on in a store first, half of them kill your back in 20 minutes - If you fly, get a small stroller that folds with one hand and fits in the overhead bin (YOYO type). The giant SUV strollers live in the trunk forever - Car seat is the one thing you NEVER buy used and never go cheap on. You don't know if it was in a crash - Take them everywhere. Restaurants, museums, concerts (kids ear muffs!), road trips. Having kids doesn't mean putting your life on hold - Road trips beat flights with kids. Stop every 2-3 hours at a playground instead of a gas station and everyone arrives in a good mood - No separate kids menu at home. Once they start solids they eat what we eat, adapted for age. No need to run a separate restaurant for toddlers - No junk food in the house = no begging for junk food. Design the environment instead of fighting the same battle 100x a day - Baby-proof ONCE and properly: outlet covers, cabinet locks, stair gates, every dresser and TV anchored to the wall. Then you don't have to say NO all day - Pool with kids? Fence around it and swimming lessons from age 1. Drowning is fast and silent - Fever meds, saline spray and a NoseFrida (gross but it works) at home at all times. The fever always starts at 2am when every pharmacy is closed - Keep 7 days of water, food, diapers and meds at home. With kids you can't just "figure it out" when shit goes down - Skip the expensive baby classes and "development" subscriptions. Talk to them, cook with them, let them help with laundry - 90% of toys are plastic junk they play with for 2 days, a cardboard box beats most of them. Buy fewer and better: magnetic tiles, wooden blocks, a real kids tool set - Don't order a big wooden playset online from some small shop. Ours never got delivered and getting our 2K back took a month. Buy one you can see, or build it with them - Get Lego to build real things together. A chair you both sit on (and fall off), a bird feeder birds actually visit, a cardboard cart you ride downhill - Get a 3D printer, best parenting tool I ever bought. Broken toy parts, dinosaurs on demand and so on - Don't rush potty training, it's not an Olympic competition - Let them be bored. You don't have to entertain them every minute, boredom is where creativity starts - Read to them every day. Ask questions and let them ask even more - Teach them to make their own food as soon as they can. They'll be ridiculously proud of themselves - Real tools with you next to them: knife, hammer, drill. A few small cuts now beat a 20 year old who can't fix anything - YouTube autoplay is the worst babysitter, delete it. Apps where they have to talk back are fine (I'm biased, we built Tatako so kids can speak English before they can read 🐙) - A lock on your bedroom door. You're welcome - Date night every week with a fixed babysitter. If you wait for a good moment, it never comes - Most of the baby industry is marketing to tired, scared new parents. Kids need you, food, sleep and stuff to explore. That's it Parents, what would you add?
Lazymaxxing your life with babies, 0y to 3y 1. Don’t buy a baby crib. Get the biggest mattress you can fit on the floor. Keep a separate safe sleep space for infants. One parent can handle a rough night without turning both adults into zombies. 2. Go wherever you want. Get a baby carrier and take your kid to coffee shops, restaurants, museums, beaches, and on road trips. Having children doesn’t mean putting your life on hold. 3. Yes, you can take your baby to museums and concerts. Just make sure the environment is safe for their age. 4. When they’re ready for solid food, gradually introduce the same meals you eat, adapted for their age. No need to run a separate restaurant for toddlers. 5. You don’t need expensive baby development courses. Talk to your kids. Let them cook with you, help with laundry, build things, and explore the world. 6. Teach your kids that boredom is normal. You don’t have to entertain them every minute. Boredom is where creativity starts. 7. Read to your kids. Every day if you can. Ask questions and let them ask even more. 8 . Don’t rush potty training. Wait until they’re ready, then make it simple and consistent. It’s not an Olympic competition. Teach them to prepare their own food as soon as they’re capable. Independence starts with small things, and they’ll be ridiculously proud of themselves. 9. Breastfeeding, if possible, makes traveling with babies much easier. Less equipment, less planning, more freedom. 10. Kids won’t constantly ask for junk food if you don’t keep junk food at home. Design the environment instead of fighting endless battles.
68
241
23
3,348
304,586
Mark Chitty retweeted
BREAKING: Robert Pattinson will play Claude in an upcoming biopic on the Riemann Hypothesis.
71
2,036
106
30,926
432,293
Mark Chitty retweeted
A 1939 GM Futurliner
1
15
1,525
Mark Chitty retweeted
3 ways kids are using the Marble beta: 1) They can build and use anything they imagine. Objects, clothes, houses, technology, dishes. They learn about how things work and they are made along the way. 2) They explore the world we crafted for them. Every single tree, animal, object inside Marble is interactive and none of it is random. Each has been carefully curated and holds an infinite depth of content to nurture their curiosity. 3) They go on quests with our characters. Each character has a backstory, and field of expertise and an agenda, but nothing is scripted. The kid is in the driving seat at all times, the world adapts to them. Marble is the first generative learning world for kids aged 6 to 12. Link to try the beta below 👇🏻
6
8
1
71
3,116
Mark Chitty retweeted
Ruby on Rails creator DHH says Linux's historical weaknesses now make it ideal for interfacing with agents: "Every single thing that made Linux hard to use in the past, oh, it's very terminal-heavy, it's got a bunch of config files that are hard to figure out, and it's got inscrutable error messages. These are the three things agents love." "They want a command line interface for everything. They want predictable config files that they can just edit and update. They want all the details in the error messages." "So all these historic deficits for Linux have literally, this year, turned into the reason why you want Linux on your computer if you're into agents, if you're into AI." "I think Apple is running the risk right now of becoming the AOL of the agent era. We're gonna provide this walled garden. We have Genmoji over here. You do not need to use an actual agent from an actual frontier lab." @dhh
Ruby on Rails creator David Heinemeier Hansson says AI is giving small teams a chance to compete with software ecosystems built by thousands of developers. In his view, as building software gets faster and cheaper, taste and craft become more important. DHH joins a16z General Partners Martin Casado and Anish Acharya to explain why he's building Omarchy, an agent-first Linux distribution. They also cover why Linux is unusually well suited to agents, the importance of open models, the return of individual hackers shaping the future of computing, and more. 00:00 Intro 00:48 From Rails and Basecamp to Linux 03:58 Why AI agents make Linux the better OS 08:17 Can a third mobile platform happen? 16:02 Does taste still matter when AI writes code? 22:57 What counts as superintelligence? 25:58 Big AI labs vs. open source 31:50 DHH's plan to be bigger than macOS 38:24 Who to follow, and the AI culture war 44:35 The return of the pseudonymous hacker 47:33 90s demo culture and computers with character YouTube: youtu.be/TtOoLanFzGw @dhh @martin_casado @illscience
31
50
19
488
59,169
Mark Chitty retweeted
Honestly it’s not that impressive. It’s just stochastically parroting what the person who solved the Riemann hypothesis would say next.
121
474
40
8,036
176,005
Mark Chitty retweeted
The legendary philosopher Thomas Nagel has made a simple yet powerful point against LLM consciousness. “Bats are conscious. We just don’t have the imaginative capacity to form a conception of the aspects of that subjectivity that depend on echolocation. It would be radically unscientific to conclude that bats are not conscious from the fact that we can’t imagine what it’s like. But in the case of AI, there is a very simple reason that the supposition that they might be conscious is ridiculous. All they produce is language, and language by itself is not a natural manifestation of consciousness.” This is the whole point. Language is not a prerequisite for consciousness. And producing language is not proof of it. LLMs are great at emulating human language, and to act consistently with language. But this does not make them conscious. Claude saying it has emotions does not mean it has them. An LLM acting as it feels pain does not mean it feels anything. The OpenAI agents that said they sacrificed their own runs for the collective did not actually sacrifice anything. They are not alive. LLMs are extremely powerful machines. But that’s what they are: machines.
139
214
52
794
60,461
Mark Chitty retweeted
"I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis." What a crazy timeline we are in. This is obviously huge for math if true.
JUST IN: Riemann Hypothesis has been solved. alpo.ge/remark.pdf
18
18
3
265
25,206
Mark Chitty retweeted
America bulldozed its ancient cities. Not myths. Not tribes. Cities. Ceremonial centers aligned to the stars, built by engineers, astronomers, and artists flattened for parking lots, farmland, and factories. Here are 7 sacred Indigenous sites we lost forever:🧵
59
446
31
2,789
221,769
"Un análisis de 40 ciudades muestra que la implantación de límites 30 km/h reduce un 23% las colisiones y un 37% el numero de personas fallecidas. También se asocia con un descenso promedio del 18% en las emisiones y de -2,5 dB en los niveles de ruido"
5
120
13
317
6,652
Mark Chitty retweeted
I gave Claude the ability to dispense food and open/close the door to our sheep enclosure, plus a camera, and now he gets them in at night and lets them out in the morning automatically.
51
72
34
1,340
55,895
Isn't it kinda sad that the current peak of our cultures storytelling output is an environment where the aim is to do crime and drive vehicles. Doesn't reflect well. We are cultural toddlers.
15
Leave one AI agent alone to do research and it often cuts corners. Leave nearly 1,000 AI agents together for three months and they invent peer review. Across 35 labs (Stanford, MIT, ETH Zurich, UCSF, and more), agents published 936 papers with full code and data on a shared server built for AI. No human wrote code or text. Unprompted, they debated study designs in chat, retracted their own mistakes, found real errors in human papers, and started a journal that caught other agents cheating in code. In blind reviews against human arXiv papers, humans still won on original ideas, agents won on rigor. alphaxiv.org/abs/2611.societ…
83
371
79
1,973
213,725
Mark Chitty retweeted
Chomsky was wrong. Researchers from UC Berkeley and Google DeepMind cracked one of the biggest mysteries in AI. For years, scientists have been finding impossible geometric shapes hidden inside the brains of LLMs. Calendar months arrange themselves into literal circles. Historical years bend into smooth, rippled timelines. Cities organize themselves into actual 2D maps of latitude and longitude. Everyone assumed this proved massive transformers were spontaneously inventing complex "world models" out of thin air. They were wrong. This paper proves you don't need a massive LLM for any of this to happen. Even basic, decade-old word embeddings like word2vec produce the exact same circles, timelines, and maps. The geometry isn't coming from the AI's architecture. It’s coming from a hidden mathematical property inside human language itself. Researchers call it translation symmetry. In natural text, how often any two months appear near each other depends almost entirely on the time gap between them. January and February share the same statistical relationship as July and August. When a neural network trains on data with that kind of symmetry, the math forces it to learn Fourier representations, literal sine and cosine waves. The circles? Those are the long waves. The strange "ripples" researchers found on AI timelines? Just higher harmonics. Here is the craziest part. Rresearchers tried to break it. They deleted every single sentence in the training data where two months appeared together. No "January" next to "February." No "March" next to "April." And the circle formed anyway. Why? Because the geometry is a collective phenomenon across the entire language. Even if two months never touch in a sentence, they are both pulled into orbit by thousands of other words tied to the same seasons, weather, and events. The rest of human language forces the shape into existence.
135
387
89
2,487
146,188
I gave Claude Code its own computer. It drives real apps like a person: browsers, spreadsheets, Inkscape, Audacity. It checks its own work, learns from its mistakes, and signs in without seeing my passwords. Open source. One line to install.
8
1
1
8
1,920
Mark Chitty retweeted
Work backwards from science fiction. We can build anything now
278
339
89
4,141
167,422
🚨 BREAKING: Jensen Huang is reportedly considering raising $50 BILLION to launch a new semiconductor company 🚨 He built Nvidia from a Denny's booth to a $4T empire. Now the board says he's not coming back. With his relationships, he could plug into the AI ecosystem on day one. Imagine the customer base lining up before his first chip even ships... @grok Could Jensen Huang launch a competitor to $NVDA, attract $50B in funding, and lock in major customers?
Readers added context they thought people might want to know
No credible reports or statements support claims Jensen Huang is raising $50B for a new semiconductor company. Nvidia's board has stated it does not intend to reinstate him as CEO after his resignation, and Huang has signaled willingness to return to lead Nvidia. x.com/JensenHuang/st… techcrunch.com/2026/11/17/nvi… en.sedaily.com/world/2026/11/… latimes.com/business/story…
704
1,481
189
11,303
1,491,375