Stephen De Winter 🤘 (@stdewinter) header
Stephen De Winter 🤘 (@stdewinter) profile picture

Proyectos online.

Santander, España
Joined January 2009
Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)Photo by Stephen De Winter 🤘 (@stdewinter)
Stephen De Winter 🤘 retweeted
Okay, and now the model card I've been waiting all week to build and share. This has been the question on my mind since Dev Day on Tuesday, so I'm guessing it's on other people's minds as well. GPT-6.1 Sol vs. Astra vs. Opus 5.5, across every effort level on @VulcanBench Frontier v4. And before I share my own insights, one note I think more benchmarks should be sharing. What accuracy score do you really need a model/effort to hit, for your routine daily coding tasks? So many benchmarks show performance at Max effort, but I never use Max effort, never use Extra High, sometimes use High, but mostly use Medium and Low with the current frontier because that's more than powerful enough for most routine daily coding tasks. And, we all know you don't need to use the best frontier model with the highest accuracy score. Sure you can do that, but you'll spend a fortune, and you'll find yourself in the 2025 agentic coding pattern where you give a model a task and just sit and wait. On VulcanBench Frontier v4, I typically say that a model/effort combo between 80 - 85 is great for easy/medium difficulty tasks, 85 - 90 is a good choice for medium/hard tasks, and above 90, well, you may never need. For me personally, I think I need model/effort combos that score above 90 maybe 5% of the time, the sweet spot for most of my routine daily coding work is model/effort combos that score in the low 80's on Frontier v4. Do with this info what you will, we're all learning together too so take it with a grain of salt. And now, for the benchmark I've been waiting to share, see below. Live long and benchmark 🖖
31
9
4
153
13,236
Stephen De Winter 🤘 retweeted
This is also basically how everyone used the internet 20 years ago You’d end up on some guy’s website about Civil War cannons, click his links page, discover a county historical society, then spend 40 minutes reading a forum for antique tractor owners “What sort of website is this?” was half the experience We called it surfing for a reason, now you open the same 4 apps every day
Noticing that old people tend to explore their environment like they’re playing some real life version of Skyrim. Just wandering into random buildings and opening a dialogue tree. “What sort of business is this?”
154
1,665
51
21,160
813,003
Stephen De Winter 🤘 retweeted
𝗜 𝗯𝗿𝗼𝗸𝗲 𝗱𝗼𝘄𝗻 𝘁𝗵𝗲 𝗲𝗻𝘁𝗶𝗿𝗲 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁 𝗺𝗲𝗺𝗼𝗿𝘆 𝗹𝗮𝘆𝗲𝗿: 𝗵𝗼𝘄 𝘁𝗼 𝘀𝘁𝗼𝗿𝗲 𝗶𝘁, 𝗳𝗶𝗻𝗱 𝗶𝘁, 𝗮𝗻𝗱 𝗸𝗲𝗲𝗽 𝗶𝘁 𝗰𝗹𝗲𝗮𝗻, 𝗶𝗻 𝗿𝗲𝗮𝗹 𝗰𝗼𝗱𝗲. Free 30 min course on AI agent harness memories. Whiteboard first, then SQLite vs @mem0ai vs @zep_ai vs LangMem on the same facts. 🌟 Github Repo: github.com/ShenSeanChen/Waku… 💻 Join our Community: seanchen.io Most people bolt a vector database onto a chatbot and call it memory. The actual layer is four operations: Store → Retrieve → Consolidate → Retire @NousResearch Hermes and my own Waku Agent run zero embeddings. Plain text and one SQLite file. Open it, it's yours. Retire is the one nobody implements. Deleting a fact and invalidating a fact are not the same system. Full video in the first reply 👇 0:00 - Why every LLM call starts with amnesia 2:31 - Store memory as text, tables, or graphs 5:04 - Find it 4 ways: do nothing, keyword, RAG, Graph RAG 7:29 - Maintain it: add, delete, retire, attribute, reflect 9:47 - Build the whole thing in plain text and SQLite 13:13 - Add a vector store only when you actually need one 14:38 - Row memory vs graph memory vs temporal graph 21:03 - Run the code for all 5 You Can Build Anything. You Can Learn Anything. 💪
23
53
3
404
25,107
Stephen De Winter 🤘 retweeted
GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday. We’re expanding preview access globally now.
2,414
6,574
4,081
46,784
9,979,376
Stephen De Winter 🤘 retweeted
can't believe i spent my whole life becoming Good At Computer only for Computer to become Better At Computer
182
2,068
73
35,260
744,015
Stephen De Winter 🤘 retweeted
Scraping websites and having AI summarize them, so that no one visits the websites, is theft. Training AI on videos, so that it can make new videos that compete with them, is theft. We are witnessing the largest theft of creative work in history.
201
2,549
102
12,064
332,130
Stephen De Winter 🤘 retweeted
Search is full of ads and wrong answers. Every other email is an ad. Prime Video charges you and shows ads. Paramount? Ads. Peacock? YouTube? Hulu? Ads followed by more ads. Netflix full of ads. Meta and X, every other thing is an ad. Pinterest is nothing but ads. AI is in everything. AI finishes sentences incorrectly and won’t stop. AI reads your email and search history to target you with more ads. Every time you open an app or visit a site there’s an update making it worse. In a hurry? First, click here to agree to terms you don’t have time to read and must accept. You need an account to do that. Change your temporary password. Enter your 2FA code. Check your email and enter that code. Now use a passkey. Your password is too simple to remember. Change it. No, not like that. Now log on. Enter your 2FA code. Check your email for a code… Welcome back! We’ve updated our terms of service and privacy policy (you have none). Subscribe to the site. Subscribe to Netflix. Subscribe to toilet paper. Subscribe to these groceries. Pay a membership fee for the right to subscribe then tip your driver who delivers the subscriptions your membership lets you subscribe to. Time to work? We’ve got to update your laptop and will slow down everything you do until you agree to update. But first, click here to agree. Update installed — your laptop’s broken now. It doesn’t matter, since your boss just replaced you with AI. Go to your phone to complain on social media. Wait, your phone needs an update so we can add more AI. Click here. Oh sorry, your phone can’t handle this update. Now it’s useless. Go get the newest phone. Here’s a text from a friend, an email, a voice mail they left three days ago but you didn’t see until now because of sync problems with the cloud. It’s their GoFundMe. Their MLM. Their Patreon. Never mind, you didn’t respond to their text within 9 minutes and now you’re no longer friends. They blocked you. Make new friends. Download this app to find people in your area. In your neighborhood. On your street. Two doors down from you. Do you know this person yet, we think you’d get along. You need an account to use this app. That username is taken. Enter a password. Not that one, you used it on another site. You need to be connected to WiFi to download the app. Allow the app to connect to other devices on your network. Allow the app to access your contacts, know your precise location, store your credit card details. Oops, sorry, we got hacked now all that info is available on the web. There’s a class action suit. You can join. It’ll take a decade to get your $3.73 share of the ten billion settlement. We’ll send it via PayPal or deposit it to your bank, just tell us those details. Oh no, another hack. That info is circulating now, too. Here’s a spam call, a spam email, a spam text. Why are you angry? Why are you talking about getting rid of your phone? Why don’t you like AI, it lets us make all of this easier? Do you know how ridiculous that sounds? This is progress. You’ll be left behind. Do you want to be left behind? Do you???
765
9,687
1,028
44,138
4,324,240
Stephen De Winter 🤘 retweeted
Best examples of brands doing AI animation/claymation/pixar style ads?
12
1
54
9,499
Stephen De Winter 🤘 retweeted
Do people realize that Mistral AI has the same revenue as xAI?
Both Mistral AI and xAI are private companies with limited public financials. Based on latest reports: - Mistral AI: Annualized revenue over $400M (as of Feb 2026, up 20x YoY). - xAI: Q3 2025 revenue $107M (nearly doubled from prior quarter), implying ~$400M+ annualized if trends continue. They're comparable in scale, though xAI shows rapid growth amid high spending. Sources: Reuters, Tech in Asia.
3
5
1,512
Stephen De Winter 🤘 retweeted
This paper shows you can predict real purchase intent (90% accuracy) by asking an LLM to impersonate a customer with a demographic profile, giving it a product & having it give impressions, which another AI rates. No fine-tuning or training & beats classic ML methods. This is BEYOND insane:
Readers added context they thought people might want to know
The paper reports that the method achieves ~90% of human test-retest reliability (correlation attainment) for aggregate product rankings, not 90% accuracy in predicting individual purchase intent. arxiv.org/abs/2510.08338
91
262
47
2,801
296,015
Stephen De Winter 🤘 retweeted
Estamos pervirtiendo un derecho fundamental?
10
2
3
19
7,194
Stephen De Winter 🤘 retweeted
AI lets you search by concepts instead of by keywords.
383
528
88
8,310
792,878
Stephen De Winter 🤘 retweeted
I am starting to think sycophancy is going to be a bigger problem than pure hallucination as LLMs improve. Models that won’t tell you directly when you are wrong (and justify your correctness) are ultimately more dangerous to decision-making than models that are sometimes wrong.
199
407
96
3,317
381,815
Stephen De Winter 🤘 retweeted
2025 is 50% complete.
215
7,159
1,424
35,814
2,185,266
Stephen De Winter 🤘 retweeted
🔥 The way we get a medical diagnosis has changed forever. As I mentioned a while back, I’d been dealing with pain in my right hand for over a year and a half, and still hadn’t received a proper diagnosis. The other day, I tried something as simple as taking a couple of photos and drawing with a pen on the areas of my hand that hurt and showed it to ChatGPT o3. Possible diagnosis? Cubital tunnel syndrome. My next question was how I could confirm that. It sent me a link to a doctor (yes, a specific one) in my hometown, Murcia, and suggested I get an electromyography done. Three days later, I’m at the doctor’s office. When I got there, the doctor asked me, ‘So, why are you here?’ I told him, well, my physio recommended to go and he mentioned I might have this cubital syndrome thing and that I should get an EMG (honestly, I was embarrassed to admit it was an AI that told me 😅) He ran the test and sure enough, ChatGPT’s diagnosis was spot on. I’ve got cubital tunnel syndrome. Once the doctor gave me the diagnosis and a report, I asked, ‘Okay, so now what?’ And he basically said, ‘Just rest’... Ummm, really? That's all? 😂 So I uploaded the full PDF report back to ChatGPT, now that I trust it even more, and it gave me a full weekly recovery plan for the whole year: everything from the exercises I should follow on YouTube to recommended vitamins and meds for this kind of thing. I still doublechecked with a doctor friend, and he said it was 100% perfect. And this is just one example of what’s happening every day now with AI. Not just for me, for you too. From legal advice at the level of the best lawyers, to developing any kind of software no matter how complex, generating entertainment, even helping you with tax regulation... anything. What a time to be alive!
194
322
80
3,957
575,502
Stephen De Winter 🤘 retweeted
¿Tienes planes a mediodía? Pues cancela, porque hoy tengo el enorme placer de participar en el podcast de @Lu1sma y @alejandronovoag junto con @chuisochuisez En una horita 👇 youtube.com/watch?v=AoyfWZxt…
1
9
1
33
2,528
Stephen De Winter 🤘 retweeted
Acabo de ver el comunicado de la liga en el que sacan pecho por haber cerrado cloudflare via movistar/o2 ESTOY INDIGNADO Perdí tres horas de trabajo intentando descargarme modelos de hugginface y resulta que no funcionaba porque a algún listo se le ocurrió bloquearlo
28
172
8
1,070
57,371
Stephen De Winter 🤘 retweeted
Tip: Stay away from people who complain a lot. Be around people who are actually building or creating something cool.
47
41
10
588
34,717