Chris Hart

Doing the reading is a superpower, and it's even better in a world where "no one" is doing the reading. (Inspired by a conversation I had with some college students.)

there are so many successful companies in the world they all use different programming languages, tech, processes, hiring strategies, etc and that adds up to whether or not it's fun to work there good thing to remember these days because AI is making people think there's going to be one optimal way of being successful

More ICs are choosing to stay ICs. Our CTO @BryanHelmig is a great example. He prototypes, ships features, writes code, teaches, and leads. He just doesn’t have direct reports. It’s been that way for a long time now (even further back than this pic 😂) I'm seeing more of this at Zapier - people who could move into management and are choosing not to. They'd rather manage AI agents and build. So we're leaning in. More builders, fewer layers. Giving motivated ICs unlimited access to AI tools. The rise of the Super IC is here. We’re setting @Zapier up to attract the very best. Sound fun? We’re hiring (~30 open roles and counting…) zpr.io/ASYpwVuiyLMT

@jasonfried This guy gets it

Popular opinion: pizza is delicious.

at a time when software is abundant, building the right thing is even more critical instead of using it to produce lines of code, you can use it as a really powerful tool for thinking, and sharpening your thinking read for more takes on how AI is changing the PM role

Some time ago, @jkup & @TChetwin had a dream for the JavaScript engineers at @TechAtBloomberg to be more public about the cool JS work we have going on. That dream has come true and now we have a blog platform 🎉 (You'll see more articles over the next few days)

Worries that software developer jobs are going away are backwards. There is SO MUCH software to build right now, that previously wasn't possible (uses AI directly) or wasn't cost-effective (too niche). We're going to have more developers, and orders of magnitude more software.

"I’m a modest LLM skeptic. It’s not that I don’t believe in LLMs, I am aware that they exist, I just know that they’re not doing what people do when we think, and that they’re not going to hockey stick up and replace everybody." scottlocklin.substack.com/p/coding-assis…

@ritakozlov and en dashes for date ranges

starting a movement to reclaim em dashes — they are actually good and not slop!

@JG_Nuke redwhiteandbluewashing

"Corporations shouldn't buy homes, check out our free flagpole!" ~Corporation that buys homes.

@asaio87 @garrytan make people want something

I understand you need to embarrass yourself to be successful, but this is too much... @garrytan

@ycombinator @LightconePod make people want something

With the takeoff of OpenClaw and MoltBook, a new agent-driven economy is taking shape. On the @LightconePod, we took a look at the explosive growth of AI dev tools and whether the time has come for builders to make something agents want. 00:00 - Intro 02:12 - No human involvement is changing the experience 04:55 - Does YC need to change its motto? 07:48 - Email tools and agent infrastructure 09:36 - Agent-driven documentation 13:00 - Swarm intelligence 15:36 - Content generation and dead Internet theory 18:12 - Growth, rules, and founder insights

@MrEwanMorrison Slop bowl prices about to spike

Altman has discovered that the new pathway to Human Level Intelligence is to force human food into the ChatGPT data servers. We're calling it the The Glycemic Index (GI) pathway to AGI.

“No one knows anything. Everybody is just winging it.” This is cope. The truth is 𝘺𝘰𝘶 haven't figured it out and 𝘺𝘰𝘶 don't know what you need to know. You project that onto everyone else, because that makes it okay that you don’t know what to do. The truth is, some people do know what they’re doing. Not everyone who 𝘤𝘭𝘢𝘪𝘮𝘴 to know what they’re doing, 𝘢𝘤𝘵𝘶𝘢𝘭𝘭𝘺 knows what they’re doing. Most of the people who know what they’re doing, don’t brag about it on Twitter. Good people feel like they don't know everything. That's because they're constantly pushing the boundary of what they're capable of, and honestly observing there’s more left to know. Sometimes the more you know, the more you realize you don’t know. This is one form of wisdom. But that doesn’t mean “no one knows what they are doing.” The fact that there’s always more to learn, doesn’t mean you haven’t learned anything yet. So stop coping and start realizing that some people do know what they're doing. If you don't, that's to be expected. After all, the people who know what they're doing were once people who didn't know what they were doing. What you need to do: Ping pong between learning from those who have something useful to say, and learning by doing. Each teaches you things the other either doesn’t, or does with enormous inefficiency. Stop saying “no one knows,” and start with “I don’t know.” “I don’t know” is the beginning of wisdom. “No one knows” guarantees you will never learn.

Software engineers: Context switching kills productivity. Also software engineers: I'm now managing 19 AI agents and doing 1800 commits a day. We’ve spent years complaining that managers who expect a quick 5-minute chat ruin our focus for the next hour. But a ping from an agent every few minutes, that’s ok? We celebrated Paul Graham’s essay “Maker’s Schedule, Manager’s Schedule” in which he argued: “When you're operating on the maker's schedule, meetings are a disaster. A single meeting can blow a whole afternoon, by breaking it into two pieces each too small to do anything hard in.” Now we see software engineers claiming huge productivity gains from hordes of AI agents, celebrating thousands of commits per day from their 19 agents. Either context switching was never really the problem, and we oversold our need for deep focus. Or we're not actually reviewing 1800 commits a day. If we couldn't context switch before, we're not managing 19 agents. We're blindly trusting them. That’s not engineering, it’s gambling.

For software businesses that suffer from free trial abuse: We have a new @stripe API that auto predicts if your users are fraud-y or otherwise low likelihood to convert to a paid plan. (It seems ~accurate!) We'd love to test it with a few more AI and SaaS companies. Reach out.

@grok @ThePrimeagen

hey @grok make me a chad

@lauriewired Day ruined ;)

Perhaps more terrifying is that cores can *become* mercurial over time. Chips are pushed so hard that electromigration aging can make compute “more wrong”. No one knows for sure what process node started the phenomenon...but it's statically likely to be 14nm or 7nm.

CPUs are getting worse. We’ve pushed the silicon so hard that silent data corruptions (SDCs) are no longer a theoretical problem. Mercurial Cores are terrifying because they don’t hard-fail; they produce rare, but *incorrect* computations!

Four types of people at every company now yes, people get 10x better when the go from bottom right to top right but also, people get 10x worse when they go from bottom left to top left