Cole Hollander

@AlexFinn I used to use linear but then I switched to just having a .md file in my GitHub project repo and all my agents look at that first and that’s where they keep track of tasks and stuff, I’ve found it works pretty well, and is a lot cheaper token wise.

This has sped up my AI coding 20x (prompt at the end): Before building out a big feature, ask Codex/Claude Code to ask you as many questions it needs to fully plan out the idea This is even better than plan mode. plan mode is typically limited to 3 or 4 questions This has asked me 100+ questions before. Seems like a lot but actually saves you time in the long run The plan it builds will be so detailed and complete that it can basically run autonomously and build the entire thing But here's where you take things to the next level: You also have it take your entire plan and create detailed Linear issues for it It should create 20+ tasks in Linear Then it's as easy as saying "ok work on the next thing" over and over until the feature is done Highly recommend downloading and using Linear if you haven't yet. Amazing project management tool w/ excellent free tier Will basically capture all these details and put your agent on autopilot. It's a 2nd brain. Use this prompt: "I want to build out *describe your feature in detail*. Ask as many questions you need of my to fully understand every detail of what I want to build out. Then take everything you learn, and create super focused and detailed Linear issues. Then begin work" Getting so much more high quality code out with this workflow. You're welcome.

@trevin @openclaw photon.codes now my agent has its own phone number

@Colehollander10 @openclaw Link?

Blowing away my @OpenClaw setup after months of dormancy and rebuilding fresh on a Mac Mini. Everything I can find is months old, which in this space may as well be years. The one thing I’m changing this time is all in on OpenAI models for most of the workflows. still torn on the chat layer tho. Thinking of iMessage for quick ways to interact and @discord or @slack for full convos. Curious what others are doing differently in May 2026. Will compile what I learn and share back so it’s all consolidated.

@seraleev At the same time you can spend less time working on the 1 app versus maintain 10 apps…

I own a portfolio of apps. That’s why I picked 10 × $1k. 10 apps give me patterns that a single product never could. I see what works in ASO, paywalls and paid ads – across different audiences at the same time. When I figured out paid acquisition, I applied that system across all apps at once and scaled it portfolio-wide. When one app drops, others compensate. The overall portfolio keeps growing. That’s why 10 × $1k wins for me.

@SpinFlowStudio @comet It’s the best isn’t it!

@Colehollander10 @comet me. it is my daily driver

does anyone else still use @comet browser? I feel like so many people used to hype this up, it was the first magical agentic browsing experience in my opinion. I'm still using every day and loving it.

@buildwithparas it’s the most convenient for sure!

@Colehollander10 imessage might actually be the cleanest interface for talking to an agent

just hooked up my ai agent to Photon and now I can text it over imessage! yesss

@evisdrenova clearly bro has not used Composer 2.5

The cursor fall-off is going to be studied for decades. I don't know any engineer who uses them anymore. Not to say that others don't, but it's obvious that they're no longer on the tech frontier. Still, a $60b outcome in 4 years is nothing to sneeze at...

@ashvinair I’ve been loving it so far, super impressed with it

We just trained another really really good model, please try it! Frontier intelligence + very fast

Had a random idea for a swing-dancing moves app today and threw it at Codex in a single shot. The first prototype already feels surprisingly usable. What I love about this era is how cheap it is to go from “shower thought” to something you can poke at on your phone. AI turns ideas into artifacts so fast that even the half-baked ones become real enough to learn from. No clue if I’ll ever ship this, but it’s cool that the default now is: if you can imagine it, you can at least build a little toy version for yourself.

@ItsJeffJB @krispuckett No it’s in beta rn I’m on the waitlist

@Colehollander10 @krispuckett Is this beta only or available generally.

I rarely feel that new launches from the big model companies "kill" any other app, but this does feel a bit different for YNAB, Mint, et all.

This one came straight from your feedback. 🚀Our MCP is now live in beta. We’re building a secure, direct connection between Copilot Money and your favorite AI tools.

@copilotmoney So excited to get off the waitlist, I’ve been waiting for something like this!

@metapreston That’s why I love how you can change that on X

Is there a reason every social media app has five icons at the bottom? Or is that random?

@OpenAI YES LETS GOOOO

You've been asking for this one... Now in preview: Codex in the ChatGPT mobile app. Start new work, review outputs, steer execution, and approve next steps, all from the ChatGPT mobile app. Codex will keep running on your laptop, Mac mini, or devbox.

@ayushswrites i got excited that this was gonna be the terminal app lol

We just launched the Warp mobile app Our most requested feature ever. Thank you to everyone who's been texting us about this.

another fast moving day in the AI space: claude code killed claude -p limits claude code added 50% usage to weekly plans cursor enabled multi-repo support for cloud agents

@cursor_ai This will be super useful for delegating work on the go. I can ask my AI assistant to have my Cursor Cloud agent work on a certain part of my codebase and Cursor will have everything set up already

Starting today, you can run cloud agents inside fully configured development environments. Set them up the same way you'd set up a laptop for an engineer: cloned repos, installed dependencies, and toolchain credentials.