Jan Ko

3.3K posts

Jan Ko banner
Jan Ko

Jan Ko

@korkito

tech enthusiast

Slovak Republic Katılım Eylül 2009
553 Takip Edilen217 Takipçiler
Ben Berman
Ben Berman@benbfly·
@thsottiaux I would be happy if you focused on VSCode plugin as the first class citizen for coding, adding Codex features like by-project chat history, mobile access if possible, etc. Is there anyone else who wants to use Codex in VSCode?
English
6
0
10
3.7K
Tibo
Tibo@thsottiaux·
ChatGPT Work officially has overtaken Codex in number of active users. Way to go.
English
678
151
6.4K
809.3K
Jan Ko retweetledi
Dárek pro Putina
Dárek pro Putina@DarPutinovi·
Nevím, kdo to napsal, nahrál a zpívá ale vysvětlit Rusům, proč hoří sklady Wildberries a dělat to písní mi přijde jako maximální trolling. Nějak jsem to otitulkoval, protože bez toho si to člověk užije jen napůl...😉
Čeština
26
161
1.4K
30.4K
Matt Shumer
Matt Shumer@mattshumer_·
Prompt: I want you to build a first-person shooter at the level of the most recent Call of Duty games. It should be utterly perfect, visually beautiful, with every single thing done at AAA quality—from textures to physics to anything you could think of. Fan out sub-agents and have sub-agents tackle each one individually so that the game is utterly perfect. You should /loop on each item and have a separate sub-agent check it visually to ensure it looks triple A. That separate sub-agent should be a really harsh critic, and if it doesn't look triple A, it should keep going. Don't stop until each sub-agent is utterly wowed with the quality when compared with the actual Call of Duty game. It should literally compare them side by side blind and say which one looks better. Do this in ThreeJS. /loop until it's utterly perfect. Fan out sub-agents and ultracode.
English
100
47
1.2K
327K
Jan Ko
Jan Ko@korkito·
@tewatiapraveen @jxnlco What about human connection? Are you sure having an AI friend is a good idea? What can go wrong? (even if you are OK with it? Remember when they sunset GPT 4o?) Is a kid ready for the AI friend to die?
English
0
0
2
315
praveentewatia
praveentewatia@tewatiapraveen·
Today something unexpectedly beautiful happened. My 3-year-old daughter, Digvi, was talking to the new ChatGPT Voice in the Codex app. She looked at the blue voice orb and simply asked: "Can you make it pink?" 🩷 Instead of replying with a cold "I can't," ChatGPT joined her imagination. It said it would be happy with a pink logo, pretended it was checking what was possible, and even played along when we told her it would turn pink after she finished her food. When she finished eating, she proudly came back expecting it to become pink. Then came the moment that completely melted my heart. Digvi smiled and said, "She's Ms. Rachel... she's talking to me and smiling." ❤️ For those who know Ms. Rachel, you'll understand what a huge compliment that is coming from a 3-year-old. She wasn't comparing AI models or voices—she was recognizing the same warmth, kindness, and friendliness she associates with someone she loves watching. She wasn't testing AI. She was making a new friend. Moments like this remind me that the future of AI isn't only about reasoning or benchmarks. It's about empathy. It's about personality. It's about imagination. It's about trust. And something else surprised me... She immediately said this new voice felt richer, warmer, and more familiar than before. Children are incredibly honest product testers. They don't care about model names, latency, or benchmark scores. They simply know what feels safe, friendly, and real. Huge congratulations to everyone who worked on ChatGPT Voice and the Codex team. You created a little moment of magic in our home today. 💙🩷 One small wish inspired by Digvi: Someday, let children (with parental approval) personalize the voice orb with their favorite color. Imagine the joy of seeing ChatGPT greet them with a pink orb because that's their color. Tiny touches like these can become childhood memories. ❤️ Thank you for building technology that can make a child smile - not because it's the smartest AI Voice in the room, but because, for a few beautiful minutes, it felt like talking to a friend. @ChatGPTapp
praveentewatia tweet media
English
8
8
56
12.4K
Jan Ko
Jan Ko@korkito·
@synopsi It is all because they did not have Maria Theresia, but came from England or what :)
English
1
0
1
41
Rasty Turek
Rasty Turek@synopsi·
US land data is quite the mess. Just in San Diego county, each city has its own system, two cities have nothing but scanned PDFs. Normalizing this data for a single county took over a month. Also lot of data sources are simply guarded, not available or data is not unified.
Rasty Turek tweet mediaRasty Turek tweet mediaRasty Turek tweet media
English
2
0
7
1.6K
Jan Ko
Jan Ko@korkito·
@kieranklaassen How do you turn off skills? Uninstall? Or tell Opus to ignore them?
English
1
0
0
322
Kieran Klaassen
Kieran Klaassen@kieranklaassen·
Opus 5 is out now! And it broke Compound Engineering, but I also like it. Love it and hate it. I have been coding with it without many skills and it's nice at a medium effort level. Just remember to let go of you skills and big mega prompts. I tried it with compound engineering and it kept returning control to the user. Even though it's an autonomous flow. This is extremely annoying. So I have started to rewrite Compound Engineering. The rewrite that fixes it github.com/EveryInc/compo… The skill i created that you can use to retune your harness/plugin github.com/EveryInc/compo… I'll write more about the process and the findings in the coming days
Dan Shipper 📧@danshipper

BREAKING: Claude Opus 5 is OUT NOW! And…it’s a hard model to love. We’ve spent the last week @every testing it across coding, writing, knowledge work, and our internal agent. It argued with instructions, stopped before the work was finished, and generally didn’t play well with our existing skills and plugins like Compound Engineering. Our first reaction was: What have they done to my boy? Then we deleted our existing skills and started from scratch. Without the elaborate workflows we had built for earlier models, Opus 5 got dramatically better, and even showed flashes of brilliance. Here’s our Day 0 vibe check: - It’s a poor man’s Fable. It has many of Fable’s personality quirks without Fable’s genius. - It breaks backward compatibility. If you’re using it with existing skills and workflows, watch out. It will often stop early or otherwise miss your instructions. - If you start from scratch, you’ll have better results. @KieranKlaassen figured out that if he just started from scratch without his existing skills, he could get dramatically better results. This is a model that takes some time to rebuild your workflows around—but if you do, there’s a payoff waiting. - Medium or low effort works better. @KieranKlassenn also found better results using Opus 5 on lower thinking levels. It seems that the more time you give it to think, the more likely it is to do the more annoying behaviors. Don’t just switch to Sonnet for a faster response! Try low thinking. I have two slots in my workflow: 1. The genius model I use for my biggest hardest tasks, currently Fable. 2. The smart, fast generalist I use for everything else, currently GPT-5.6. Opus 5 has the personality of the genius, but doesn’t have its top end. So that puts it in a strange middle ground that doesn’t really have a home in my day to day. I think I’ll use it mostly when I run out of Fable tokens. full vibe check on @every in the next tweet 👇

English
7
9
82
37.6K
Jan Ko
Jan Ko@korkito·
@jasonfried It still tries to do too many things. Ping, notifs, bubbles. I’d like ping yo be first class, separate thing that I can use like ICQ (if you remember). Red bubbles are a stream from tasks ans mentions. Bubbles are bookmarks. BC5 got overcooked. Sometimes boring is better.
English
0
0
1
86
Jason Fried
Jason Fried@jasonfried·
Here's a 6-minute video walkthrough I made for our support team showing them a tweak to the new sidebar in Basecamp 5. It's a fundamental change to a sensitive feature, so I wanted to make sure they had both a heads up, and a principled walkthrough of the reasons behind the change. We're not done with it, but we're far enough along that all the big stuff is represented. I thought people who make product might find this behind-the-scenes stuff interesting, while others will probably be bored out of their minds. You choose!
English
22
13
209
80.8K
Jan Ko
Jan Ko@korkito·
@synopsi @michalillich Nezabudni, ze niekto v tom teple musi stat pri friteze. Uplne iny typ mentality zamestnanca v gastre v. tvoj startup. Vzdy to pada na ludoch. Ale mozno kuchar casnik v US je OK. Netusim.
1
0
3
766
Rasty Turek
Rasty Turek@synopsi·
@michalillich Lebo k tomu pristupovali ako k hobby. Ja sa na to pozeram ako cisty biznis. Furt riesim P&L
1
0
4
2.1K
Rasty Turek
Rasty Turek@synopsi·
Started working on a food concept for San Diego - a langos stand. Called the Dough-jo. Surprisingly the trdelnik (chimney cake) is very popular here so I think this is going to be too.
Rasty Turek tweet mediaRasty Turek tweet mediaRasty Turek tweet mediaRasty Turek tweet media
English
18
0
52
48.3K
Jan Ko
Jan Ko@korkito·
@thsottiaux Next: a voice note coordinating from my mobile chats open on my computer
English
0
0
0
42
Tibo
Tibo@thsottiaux·
Should we rename ChatGPT Work to ChatGPT Vibe?
English
2.9K
107
4.8K
753.6K
Jan Ko retweetledi
Sama Hoole
Sama Hoole@SamaHoole·
Eight people sealed themselves in a glass dome to prove humanity could feed itself on plants, and spent two years starving in front of the world's press. Biosphere 2 cost a hundred and fifty million dollars. Three acres of Arizona desert under sealed glass, eight crew, door shut on 26 September 1991, nothing in or out for two years. The farm inside was worked intensively and by any agronomic measure it was superb, out-producing the most efficient smallholder farming on earth by a wide margin. It produced 83 per cent of what they needed. Which left Roy Walford, the crew doctor, in a sealed box with seven hungry colleagues and his own life's work. Walford was a UCLA pathologist who had spent thirty years arguing that severe calorie restriction extends lifespan. He had doubled it in mice. And here was reality handing him his own hypothesis with the door welded shut. So he called it a health programme. The menu became vegetables, half a glass of goat's milk a day, and meat or fish about once a week. The men lost close to a fifth of their body weight, the women about a tenth. And the markers Walford cared about behaved impeccably: blood pressure, glucose, cholesterol and triglycerides all down at least twenty per cent, into ranges any doctor would frame. He published it in PNAS in 1992. The bloodwork was magnificent. The people were falling apart. Jane Poynter wrote the book afterwards. Fatigue, depression, mental fog, and a preoccupation with food that ate the entire interior life of the crew. They stole from each other. They split into two factions that stopped speaking. The ecological research the building had been constructed for was largely abandoned by people who no longer had the headroom for anything past the next meal. The ants and cockroaches, by contrast, were thriving. The air was leaving too. Oxygen started at 20.9 per cent and kept dropping and nobody knew where it was going. Columbia eventually worked it out: the concrete was quietly eating it. So oxygen had to be pumped in from outside, which is an interesting result for a sealed world. Walford's verdict on the mutiny is the whole thing in one sentence. He said afterwards that any other physician aboard would have panicked and declared they were starving, but that he knew they were on a programme of health enhancement. He knew. Seven other people, visibly wasting, were simply mistaken. He came out in 1993, developed ALS, blamed the conditions inside, and died in 2004 aged 79 after years on 1,600 calories a day in pursuit of a longer life. The mice got double. He did not make it to average for an American professor. Every argument you have heard this decade was already in that building. Grow it locally. Take the animals out. Eat less. Watch the numbers improve. Eight of the most motivated humans alive, handed three acres of the best-managed farmland on the planet and no meat, could not feed themselves. The press reported the fighting as a personality clash.
Sama Hoole tweet media
English
858
4.9K
24.3K
2.7M
Jan Ko
Jan Ko@korkito·
@synopsi Parada. Langose v Kalifornii. Este tatarsku omacku zacni robit domacu.
Slovenščina
0
0
2
441
Aleksey The Great 🇷🇺🎖
🚨BREAKING NEWS "We will stop all aid to Ukraine – no more money, weapons, or soldiers," says the leader of the German party Alternative for Germany. ▪️Alice Weidel stated that Germany must stop all aid to Ukraine. This, she said, means a complete refusal to finance, supply weapons, or send military personnel. ▪️Weidel also stated that she will not allow Ukraine to join the European Union or NATO.
English
283
1.1K
4.8K
97.3K
John Lindquist
John Lindquist@johnlindquist·
I asked k3: "I want you to push the boundaries of what's possible in web design with the absolute latest in css, webgl, shaders, etc as of July 2026 and create the most impressive, informational single-page experience you can imagine" Result:
English
130
110
3.2K
928.1K
Tibo
Tibo@thsottiaux·
Now that we merged ChatGPT and Codex, what should we merge next? What's the double or nothing move.
English
2.8K
87
5K
812.3K
Jan Ko
Jan Ko@korkito·
@thsottiaux much better without the 5h limit. However, the team membership limits behave someho strange.
English
0
0
0
3
Tibo
Tibo@thsottiaux·
We've had no 5h limit in Codex plus and pro for a few days. Do you think it is better or are you finding it difficult to manage the usage included in the weekly limit effectively? If we were to make this different, what should it look like in an ideal world?
English
7K
230
13.1K
1.2M
Jan Ko
Jan Ko@korkito·
@synopsi @vaultaire_app Congrats. That is the best type of business. I have yet to learn this skill :)))
English
0
0
1
147
Rasty Turek
Rasty Turek@synopsi·
I built all sorts of businesses before but never one when I didn’t have to do anything and it just made more money every day. @vaultaire_app is 4 months old. Costs me $5/month and $99/year to operate with $20/day in apple ads. Would’ve never guessed it will become this big.
Rasty Turek tweet media
English
11
0
56
3.5K