Joo Tat

Was planning out my work for the year when this struck me. To do big things is to do things that have delayed but outsized gratification. And to not burn out while doing things that have delayed gratifications is to do project management well with your past and future selves.

@nurijanian This is a banger for the number of engagements it got

junior pm: i’m leading my first global alignment call and i’m fully spiraling senior pm: ok, what’s the game plan? junior pm: book the meeting, present the proposal, leave with everyone aligned senior pm: how many folks are we talking? junior pm: about 8 stakeholders spread across 4 time zones senior pm: and that’s also when you’re introducing yourself to them… for the first time? junior pm: yeah. isn’t that what the alignment meeting is for? senior pm: there's your problem junior pm: what do you mean? senior pm: alignment meetings don't create alignment junior pm: then what's the point? senior pm: they confirm alignment you already have junior pm: i don't follow senior pm: have you talked to any of these 8 people yet? junior pm: i sent them the pre-read senior pm: that's not talking junior pm: so i should... call them first? senior pm: every single one. 1:1. before the big meeting junior pm: that's 8 separate calls just to prep for 1 meeting senior pm: now you're getting it junior pm: seems like a lot senior pm: it's the whole job junior pm: what would i even say in these calls? senior pm: you're not saying. you're listening junior pm: listening to what? senior pm: their fears. their motivations. their hidden objections junior pm: why wouldn't they just say that in the meeting? senior pm: politics. pride. audience junior pm: so they'll tell me privately what they won't say publicly? senior pm: if you ask right junior pm: how do i ask right? senior pm: frame it as soliciting their expertise junior pm: "i'm new to this and want your input"? senior pm: exactly. people love feeling like the wise advisor junior pm: and then i just... shut up? senior pm: shut up and write everything down junior pm: what am i looking for? senior pm: conflicting goals between stakeholders. who wants what and why junior pm: ok say i do all this. what changes? senior pm: two things junior pm: go on senior pm: first, you adjust your proposal based on what you learned junior pm: makes sense senior pm: second, you sprinkle their exact phrases into your presentation junior pm: why? senior pm: people support what they helped create junior pm: even if they didn't technically create it senior pm: especially then junior pm: what about the difficult ones? the ones who'll disagree no matter what? senior pm: you pre-negotiate junior pm: how? senior pm: in your 1:1, you say "i know you have concerns about X. i can't fully address them. can you disagree and commit in the meeting?" junior pm: you can just ask that? senior pm: you'd be surprised how often the answer is yes junior pm: because you asked privately senior pm: because you respected them enough to ask junior pm: so the alignment meeting itself is basically... senior pm: a ceremony junior pm: confirming decisions already made senior pm: in hallways. in 1:1s. in coffee chats junior pm: the meeting before the meeting senior pm: now you're ready to run alignment

@0xwhrrari Bookmarked and saved to my home scren

Google's former CEO just said what everyone in AI already knows Building wealth is getting easier if you actually learn the tools Not by scrolling AI threads By understanding agents, Claude Code, prompts, memory, skills, MCP, and routines Save this before it disappears from your feed Everything below is free: Agent architecture langchain.com/blog?category_… Claude Code 101 anthropic.skilljar.com/claude-code-101 Claude Code in action anthropic.skilljar.com/claude-code-in… Prompt engineering platform.claude.com/docs/en/build-… Interactive prompt course github.com/anthropics/cou… Claude.md + memory code.claude.com/docs/en/claude… Skills code.claude.com/docs/en/skills MCP code.claude.com/docs/en/mcp Routines code.claude.com/docs/en/routin… Claude Code ultimate guide github.com/FlorianBruniau… Awesome Claude Code github.com/hesreallyhim/a… Anthropic academy anthropic.skilljar.com Official Claude Code docs code.claude.com/docs/en/overvi… All of this costs $0 Most people will keep asking AI one question at a time Save this and start learning the stack

@businessbarista It’s easy to get lost in building when building gets easier but isn’t the bottleneck. Saving this to home screen to remind myself from time to time.

Best way to understand what's happening in software is to understand what happened in media. With internet & platforms, the cost of creation went to zero. Supply skyrocketed. Consumers had more choice than ever before, which created a long-tail of niches, introduced a new standard for "good", and made trusted distribution the most valuable currency.

@NickADobos What are you building?

Current dev loop: Talk to phone -> app appears 1. Connect codex to Mac and iPhone 2. Make an iPhone app and deploy it to my phone 3. Open codex on phone, open voice mode and ramble for 10 minutes while using my app. Pro tip voice mode stays on even if the codex ios app is in the background 4. Once you are ready, end with: “make a prioritized list of updates and then do them, then rebuild to my phone” So all I do is test and ramble about what’s wrong and needs fixing. I basically just dictate for 10 mins while using my app and then spend 1 minute in codex sending stuff and reading. Then flip back and continue playing with my app Repeat —— For non-team projects I never read code anymore. Only prompts and chat are worth reading. Everything else is simply using the software, fully experiencing it and feeling it, and shaping the ideal into pixels and data. My ideas manifested in machines within seconds Lightning in a rock in your pocket. Pure magic

@Replit Maybe when the system detects that a way of working with the agents is inefficient (e.g. user doesn't have the vocab to precisely describe a change), the system can intervene and introduce some concepts that could help. Same principles as this: x.com/0to1000_pdEng/…

I built redditrecs.com on @Replit 2 years ago, then eventually outgrew it as the app got more complex. Recently, I jumped back in to vibe-code a new idea. First surprise: the file panel is hidden by default. Took me a few minutes just to find it 😅 That’s interesting because even when I’m vibe-coding, I still look at the file structure to build a mental model of the app. It helps me better direct AI agents when improving or debugging things. It also helps me know what I can change quickly and deterministically myself, like swapping assets, instead of asking an agent, waiting, and then having to verify the result in-app. Feels like Replit is doubling down on non-technical operators - people who may not have the time or interest to understand how their apps work even at a high level. Fair game if that’s the ICP that’s working. But it'd be cool to see more ways to gradually educate users on the key parts of their codebase. Especially for people who want to get better at building apps, or push their apps beyond a certain level of complexity.

@PascalBrokmeier @petergyang Yeah I do this too

@petergyang Symbolic link? I have skills in a global ~/Code/ai-skills and symlink those with a small script to all the global Claude codex open code etc locations. That way all these tools have my skills globally but the instructions to each repo and subfolder locally.

It's pretty annoying that Codex uses agents.md and Claude Code uses Claude.md. There should be some industry standards to this stuff?

Btw I like to have my banger bookmarks on my home screen so I rmb to revisit them (instead of rotting away in the bookmarks tab) You can do that with: apps.apple.com/us/app/resurfx…

"But that is where the real work happens. The harder part was never the answer. It is the question. What are we actually solving? What does better look like?" x.com/maybeTuring/st…

Is AI making us more productive? Or is AI a dopamine slot machine tricking us into thinking we’re more productive? Some thoughtful posts about the failure modes when using AI:

@_adishj @garrytan @motion_so Sick. MOTION

we’re launching Motion, the frontier agent for tasteful motion design. this launch video is made entirely with Motion. 👇🏽 QT + comment "MOTION" to get 1,000 free credits. tag @motion_so in any X post for a surprise. now generally available at motion [dot] so.

@stevenkplus1 Great read. To put it another way I think the key is to be intentional about what type of thinking you’re outsourcing to AI. Bookmarked and saved to my home screen to revisit from time to time

> the more efficient you are at doing the wrong thing, the wronger you become. and as a slop machine, i decided to take this personally (just kidding! great read)

@shreyas Lots to think about. Bookmarked and saved to my home screen for revisiting (Using apps.apple.com/us/app/resurfx…)

@sairahul1 Saved bookmark to my home screen so I won’t forget to revisit this App widget: apps.apple.com/us/app/resurfx…

Andrej Karpathy's advice for beginners getting into AI: "Put in 10,000 hours of work." He's right. But most builders waste the first 1,000 hours on the wrong things. They write code before understanding context windows. They build agents before understanding token limits. They ship products before understanding what models can't do. The builders who compound fastest aren't the ones who code the most. They're the ones who understood the fundamentals before touching a single line. These are the 10 concepts that make the first 1,000 hours count ↓ Bookmark this before you start.