Sam Altman says Slack has many positives, but it creates endless fake work We need an AI-native productivity suite to replace docs, slides, email, and Slack Not add-on features, but trusted agents that handle work and only escalate when needed This finally feels within reach

If you're a dev but you don't know how to use git bisect, make that the next thing you learn. https://t.co/KTOGfGZLVK

Wow my first ever post to hackernews hit the front page.

Open sourced the whole thing. Check ` https://t.co/Jghtrz0FRd` to see how it works. Customize it to build whatever you want. It's slow, expensive, and kinda dumb, but if you want something to mess around with, go nuts. https://t.co/uSiardCNyO

These feel like problems of degree, not kind. Inference improving ~10x/year. Cost heading to zero. Context windows growing. When inference hits 100ms, the question shifts from "how do we generate better code?" to "why generate code at all?"

So is this practical? No. But the capability exists. The AI can handle application logic. It's just too slow and too expensive. The problems are all performance: speed, cost, consistency, reliability.

The AI also has the memory of a goldfish. Colors drift between requests. Layouts change. It forgets what UI it generated 5 seconds ago. When it hallucinates broken SQL, that's an immediate 500 error.

But it's catastrophically slow. 30-60 seconds per request. Traditional apps: 10-100 milliseconds. That's 300-6000x slower. Cost: $0.01-0.05 per request. About 1000x more expensive than normal compute.

The experiment "worked". Forms submit. Data persists. APIs return JSON. The AI invented sensible database schemas, injection-safe SQL, REST-ish patterns, responsive layouts. All emergent from three tools and a prompt.

The goal: test if AI can replace code entirely, not just write it. Everyone's focused on Cursor, Copilot, AI that generates source files. But what if you skip that step? What if the AI just... does the thing?

I spent a few hours last weekend building a web server with zero application code. Every HTTP request goes straight to an LLM. It decides what to do in an agentic loop using three tools: database, webResponse, updateMemory. Here's what happened →

Recent Azure/AWS outages are reminders that every cloud provider is a single point of failure in your architecture.

Companies confuse technical debt with architecture debt because the symptoms look the same - delays, outages, rising costs. But the fix is completely different. Technical debt requires building work: coding, refactoring, testing. Architecture debt requires organizing work:

Learn to ship. Shipping is a skill distinct from coding. Shipping is designing, coding, QAing, story-telling, teaching, marketing, selling, pivoting, iterating… It used to be that coding dominated in importance because of coding ability scarcity. AI will push you to go further.

Remember when you spent weeks learning Claude Code's hooks, subagents, and custom tools? You thought you were just getting better at automating your dev workflow. Turns out, you were learning something way bigger. 🎯 THE PLOT TWIST: Last month, Anthropic renamed the Claude

https://t.co/sdAs1RQsPO is great.

How could this be worthwhile when AI already works great on phones? https://t.co/8Gsq0mJQoM

He says some dumb stuff sometimes but who doesn’t? I stand with @dhh and let’s build!

UI frameworks from Apple, Microsoft, and Google are hilarious. Aqua (25 years ago) was the original liquid glass. But it turned into brushed metal/skeumorphism. Aero glass flattened into Metro tiles, Material Design lost depth overnight. Now we’re back at Liquid Glass, and