For a mobile company this is just sad

Marc Andreessen on why VCs ignore cold emails — and why that’s not unfair “The way the top-end venture capitalist firms work is they’ll basically take you seriously if you come in introduced by somebody they’ve worked with before, and they won’t take you seriously if you don’t.”

The #1 mistake I made with my first app: I focused on making it useful. But users don’t come back for useful. They come back for: • Dopamine hits • Instant feedback • Tiny wins that feel big Once I understood that, retention changed overnight

This is the *exact* ASO strategy I use to consistently rank in top #10: People tell me I'm doing ASO wrong… but it works for me (and only 1 of my apps got hit by recent App Store ranking changes...) Judge for yourself: Title field "Primary Keyword: app name" to rank on both

Andrew Wilkinson Voice chat to cursor agent to generate an initial PRD, hit taskmaster MCP to generate tasks off of it, call context7 to supercharge all of the tasks, linear MCP to log all of the work, background agents to handle each task, playwright MCP to handle testing for each task, then

Andrew Wilkinson The most powerful workflow I use is Claude Code (or Cursor) paired with the Playwright MCP (github.com/microsoft/play…) which allows coding agents to navigate a browser, take screenshots, read network/console logs, etc. It's the missing link to enable autonomous front-end iteration

These 17 B2C app pains will never go out of business: – Can’t focus – Can’t sleep – Can’t lose weight – Boosting testosterone – Social anxiety – Dieting – Debt – No time – love life – Corn use – Phone addiction – Fitness motivation – Hair loss – Skin health – Fitness – Mental

Claude code is great with agents except you run out of your daily limit very quickly and need to switch to API. Kind of defeats the purpose of being able to move quickly while keeping costs down. Also seems to lock out till top of the hour rather than rolling 5 hour limits.

visualize codebase structure with diagrams

I just need Chat GPT-5 to be consistent: - Follow instructions closely, e.g. no em dashes - Render the same image with edits using the base image, or at least not cartoonize any changes I request - Not agree with me on everything and give me the truth + gaps in my thinking

i check "ai agent for x" every month this month there is about 400 last time i checked is was about 300 probably nothing

Viral app idea: 1. Clone QUITTR (quit porn addiction): $300,000/month 2. Target a different addiction Just a few ideas: → Gambling → Shopping → Coffee/caffeine → Video games

Even with rules everywhere, what is up with claude just making up "simulations", fake data and hardcoded solutions to things you actually need working... Has anyone else experienced this with Claude? Anthropic

Is it just me or is Claude Code running on stupid mode today?

To be fair some companies spend 100x that to learn the same lesson

people absolutely sleeping on email nurture after signup to activate saas customers immediately after signup welcome email 1 hour later send how to use software day 2 why we built this day 4 case study day 7 checking in do you have questions day 10 how to use killer feature

I am once again begging you not to squander your startup's announcements. You only come out of stealth once. You only launch a product once. You only raise a Series A once. Here's how not to fuck it up: