Emad Ibrahim

I'm giving away my entire EliteSaaS Starter Template for free. $1,199 value. 15+ production-ready systems: Auth, Payments, AI, Teams, Analytics, Blog, Email Marketing, and more. Next.js 15 + TypeScript + Supabase + Stripe. Ready to deploy. Details here: emadibrahim.com/free-elitesaas

I don’t agree with his tech stack Choices or decisions but we can’t deny that his stuff sells because it works not because he used the correct coding pattern or latest version of X framework or even have good UX. My guess is it all works because he has excellent distribution and marketing.

. @levelsio pioneered the “my product barely works has no craftsmanship but, hey it sells” I’m not going to argue the opposite. But having used multiple of his products never ever felt a single ounce of joy. And that won’t change. Why are people looking up to him again?

Q: How are job postings for software engineers rising rapidly despite AI agents automating coding? A: Because there’s far more code to manage than ever before. We’re already seeing a 14x YoY increase in GitHub commits, and it’s accelerating. AI has dramatically lowered the cost of writing code, so it’s now being used across far more businesses, applications, and use cases. We’re at the beginning of a massive productivity boom driven by the proliferation of bespoke software throughout the entire economy. Coding has been AI’s breakout use case this year. The fact that it’s increased demand for software engineers — rather than decreased it — should call into question the entire “AI will cause mass job loss” narrative.

xAI's Grok Skills update is another signal that reusable AI workflows are becoming product infrastructure, not prompt trivia. My take: teams should treat skills like code: versioned, reviewed, tested, and owned by the people who depend on them. infoq.com/news/2026/05/x…

Total agreement on the opt-out. Bigger worry for me as a dad is the AI chatbots now baked into school tech, kids chat about anything and parents see none of it. That's actually why I built familygpt.chat for my own kids, lets me see what they ask AI and get pinged on anything concerning. Visibility shouldn't be a luxury.

I’m not speaking here as a school board member but as a grandma that lives with my grandson who is in Kindergarten(soon to be 1st grade)I agree with these parents. I don’t want my grandson on an IPad or Chromebook in elementary school. There should be an Opt Out option, Thoughts?

totally feel this. actually built tornic.dev for this exact reason - desktop app that keeps your Claude Code + Codex subs working even when you're away. kick off tasks via Discord, get notified when done. no API fees on top, and it gets more reliable over time as it learns from past runs.

Anytime I’m not actively running agents on Codex or Claude Code with subscriptions I feel like I’m burning my money

My onboarding brings all the boys to the yard and damn right, it's better than yours 🗄️🍼 meet runcabinet.com - my open source project with 2000+ GitHub stars and real people using it. AI workspace where your files, agents, tasks all live together and let you rest while they do the work. npx cabinetai run or Mac install (windows and linux are cooking)

Agent safety is entering the dev loop. Microsoft open-sourced RAMPART for CI-friendly agent safety tests and Clarity for design assumptions before code. My take: if an agent can act, its failure modes belong in specs, PRs, and CI. microsoft.com/en-us/security…

Shipped a free contraction timer for FamilyGPT. The product work was designing for the real moment: tired, anxious, on mobile, needing one clear start/stop flow. Duration/frequency tracking, local persistence, and a careful 5-1-1 indicator. @eibrahim/shipping-a-contraction-timer-that-respects-the-moment-d8f00060dcfa" target="_blank" rel="nofollow noopener">medium.com/@eibrahim/ship…

@ByrneHobart The "got bored and moved on" arc is the lucky outcome. What bugged me with mine was zero visibility into what they were actually asking. Built familygpt.chat for it: monitor convos, alerts on bad stuff, tune AI to your family's values. Android live, iPhone soon.

An experience that taught me a lot about the addiction potential of sycophantic generative AI: one of my kids got into using ChatGPT that way, would produce multiple 20,000+-word sessions, and then... got bored of it a few months ago and switched to making games in Claude Code.

Competition is good. The real differentiator in agentic coding is task completion rate + cost. API bills add up fast. I built tornic.dev to run on Claude Code subscription (not raw API) - it learns from failures so runs improve over time. More reliable than OpenClaw for complex tasks. on top of Claude Code subscription (not API tokens) - it learns from failures so runs improve. More reliable than OpenClaw for complex tasks. using Claude Code subscription (not raw API) for exactly this - it learns from its mistakes so future runs get better. Way more reliable than OpenClaw, and the economics just make sense.

LATEST: 🤖 DeepSeek is building an agentic coding tool to rival Anthropic's Claude Code and OpenAI's Codex, per job listings posted this week.

@thepatwalls CMUX

What's the best Mac terminal app for the Claude Code / AI command line tools? I've always used iTerm but I feel like there could be some cool apps out there that make life easier.

Google's I/O dev news is less about vibe coding and more about agent infrastructure. My take: judge coding agents like systems: permissions, state, observability, handoffs. blog.google/innovation-and…

Google moving Gemini CLI into Antigravity is a useful signal: coding agents are becoming platform infrastructure, not just nicer terminals. My take: treat agents like runtime deps. Test plugins, auth, CI access, and fallback paths before cutover. developers.googleblog.com/en/an-importan…

Agent safety is becoming CI work, not a prelaunch review. Microsoft's RAMPART turns red-team findings into repeatable tests; Clarity captures assumptions before code. My take: if an agent can act, safety needs versioned tests and reviewable decisions. microsoft.com/en-us/security…

Shipped AICurate hub logo uploads: managed Spaces/CDN storage instead of pasted external image URLs. Wrote up relative keys, image normalization, cleanup, and why good infrastructure should feel boring. @eibrahim/making-hub-logos-boring-shipping-cdn-backed-uploads-in-aicurate-08397bd69d0e" target="_blank" rel="nofollow noopener">medium.com/@eibrahim/maki…

Speed matters. But specs are where leverage starts. Article: geekwire.com/2026/aws-targe…

My take: the next big skill for developers won't just be "using agents." It'll be writing clearer intent, spotting ambiguity early, and knowing when the tool is making decisions you never actually made.

AI coding tools are getting faster, but AWS Kiro's new requirements analysis feature points at the bigger issue: Most bad AI-generated code starts before code exists. Vague prompt -> vague spec -> hidden assumptions -> brittle implementation.

Shipped a small PetRoute free tool and wrote up the build: a cat age calculator that avoids the old “multiply by seven” shortcut and uses life-stage context instead. @eibrahim/shipping-a-cat-age-calculator-that-avoids-the-7x-myth-43940aef862c" target="_blank" rel="nofollow noopener">medium.com/@eibrahim/ship…

@cljack Honestly the underrated parent-AI win - the admin tax is brutal. I went the kid-side direction myself: built familygpt.chat so mine can use ChatGPT but I see the conversations and shape responses to our values. Way less worry, same learning upside.

My claudebot just texted me that my kid needs written permission to observe the fetal pig dissection tomorrow and I was like “ugh can you handle it” and it drafted the email and sent it to the right person. Amazing, I love AI, 10x reduction in my fetal pig related mental workload