building with no code

@d4m1n because the harness matters more than the model at this point. context management, diff application, file awareness. Cursor has a two-year head start on all of it. the underlying model is almost a plug.

lol Cursor is a better harness for both GPT 5.5 in Codex AND Opus 4.7 in Claude Code how is that possible?!

@NainsiDwiv50980 the mode shift is: stop asking Claude to do tasks, start asking it to be a thinking partner on the problem. "what am i missing in this approach" hits different than "write this function"

The person who built Claude Code — Boris Cherny — is redefining how developers use Claude Opus 4.7 to write and think about code. But most people are still stuck here: “write a function” “fix this bug” “explain this code” That’s beginner mode. His real idea? You don’t use Claude. You collaborate with it like a senior engineer. Here are 10 high-leverage prompts that unlock that shift:

@ParthJadhav8 the step everyone skips: what happens when the agent marks it done and it's wrong. nobody's building the review layer yet. that's the whole product gap.

This is crazy !! Cursor now has built a Kanban board where you can just drop in tasks and the agent will pick those up and complete them.

@RodmanAi the thing that keeps happening. Anthropic ships a constraint, the open source version drops in 48 hours with no constraints. happened with Claude design, happening here. the moat isn't the model anymore.

Anthropic just lit a fire 🔥 They pulled Claude Code from the $20 plan… So devs did what devs do: Rebuilt it. Open-source. Better. OpenClaude ↓ • Any model (GPT, Gemini, local) • No subscriptions • No limits • Full coding agent in your terminal But here’s the kicker: You can route tasks across models. Cheap for simple. Powerful for complex. One CLI. Infinite setups. Gatekeep tools → Get replaced. This is going to blow up.

@Bhavani_00007 the intern asking the question at the end is the whole story. the engineers who survive this are the ones who can tell when Claude's answer is wrong. that's it. that's the only skill that compounds right now.

It’s so over for junior engineers 😭 We have a new CEO at the company I’m working in, and he recently pushed everyone to adopt a full Claude-based workflow. Now everyone is using Claude for almost everything. During a meeting, he asked if anyone had questions. An intern asked whether the company would continue hiring interns or junior engineers. He said everyone currently working is safe, but they've decided to stop hiring interns and junior engineers going forward. They don’t want to spend 3–4 months training freshers anymore, and would rather hire experienced engineers who already know these tools. Are junior engineers actually getting replaced??

@Zeneca agree but the fix isn't doing it manually. it's being able to read what it built. that's the actual gap. not "should i use AI" but "do i understand the output enough to catch when it's wrong"

Another thing I've noticed when relying on AI too much is that you have far less understanding of the things you're building and working on vs when you're doing things manually Dunno if that's just me, anyone else feel the same?

@AnatoliKopadze the gap between knowing how to use Claude and not knowing is worth $18,000. right now people are just hoping they figure it out on their own. most don't.

They quoted him $18,000. He opened Claude instead. Same result. One afternoon. $20/month. The gap between knowing how to use Claude and not knowing is worth $18,000. This article below fixes that.

@AlexFinn This is really solid

Every app I vibe code has the tech stack below Easy for beginners and free to start If you've never built an app before, just paste this list into Codex or Claude Code and you'll be good to go: Web framework: NextJS Hosting: Vercel Database: Convex Auth: Clerk Payments: Stripe Styling: Tailwind AI: logic- OpenAI, creativity- Claude, cheap tasks- Gemini Flash 3 Emails: Resend Design apps with: GPT image gen 2 + Claude Design AI I use to build it all: Codex desktop app w/ GPT 5.5 Any questions let me know!

@YashHustle_22 its always been Github

Where are you storing your code? 1. GitHub 2. GitLab 3. Bitbucket 4. Local only

@Oluwaphilemon1 I don't really see the upgrade outside of it looking cool

SHE USED CLAUDE CODE TO BUILD AN UPGRADED VERSION OF OBSIDIAN NOBODY THOUGHT WAS POSSIBLE her entire X history. analyzed with topological data analysis rendered as a living knowledge graph with hundreds of nodes LLM agents. computational biology. theory of intelligence. generative art all clustered. all connected. all alive and she controls it with bare hand gestures through her MacBook camera Obsidian was always supposed to be a second brain. she just made it actually look like one 👇

@stevelauda_ really good for turning an initial sketch into something more. it'll get much better with time too.

Who is actually using Claude Design? Anyone?

@vikrambuilds what is the AI developing if you aren't guiding it?

If AI is writing 90 percent of your Code Are you still a developer..?

@vivoplt the AI wouldnt be building your product unless you told it to.

Can you call yourself a founder if your entire product was built by AI?

@TaiTechSolution @SafockAI Thanks! Not yet honestly, been mostly focused on getting the product right first. But Reddit's definitely on the radar. Any tips on how you'd approach the intro post without it feeling too salesy?

@TFisPython @SafockAI Solid launch 👏 Are you currently leveraging Reddit as part of your user acquisition strategy? Builder communities there are highly active when products are introduced the right way, they can drive consistent signups, feedback, and early adopters fast.

Hey founders 🚀 Looking to connect with people building in: 🌼 SaaS 🌼Teach 🌼 Automation 🌼 AI tools 🌼 Web apps 🌼Production Development. Drop what you’re working on👇

@tonychang430 @InsForge_dev 401 :)

CAN WE GET TO 400 FOLLOWERS TODAY. I post about start journey and want to grow with my users together! We are building @InsForge_dev Backend platform designed for agents as the primary operator

@heyblake Crowdtest.io

Fork it Drop your landing page URL I'll give 1 piece of advice to as many of you as I can

@Mugilan_SS @OpenAI That may make me switch 😭😭😭

Codex is not like claude code. if you know the limit is going to end, like last 10 to 8%, give an very long run task, and even after the limit got ever, it will continue to do the task until the task was completed. Shout out to @OpenAI team.

@josephabahx @justbyte_ definitely better at finding nuances, I've had the 20 dollar plan for a while just to check my work.

@TFisPython @justbyte_ I have kept both subscriptions for over a year and used both extensively. GPT models are better at finding nuances. Claude is a coding machine but usually overplans or sometimes confuses itself because of the volume of content it generates.

Who actually switched from Claude Code to Codex… and was it worth it?

@josephabahx @justbyte_ I’ve only heard it the other way around where Claude is the better planner. Interesting, I personally think in my limited usage of it that it’s a better operator than Claude

@justbyte_ I have and using both extensively. Claude still better at witting code generally but Codex is better at planing. I use codex to create contracts then, Claude handles implementation. Codex for review and finding issues, Claude implement. For mobile App, Codex wins

@cryptopunk7213 Woah may be time to get back on cursor

this is huge news from Cursor, they've pulled of the impossible and turned their ai-wrapper into a in-destructible moat sam altman literally called it <24hrs ago and here we are: > cursor's ai agent harness is available for anyone to build on, which means ai models are now a commodity > 1-time install and now anyone can run cursor's agent locally or via cloud. use any model (e.g. gpt 5.5) but with the added cursor harness that makes it 10X better. > its so good that 3 of cursors biggest competitors are embedding it into their products. > now cursor DOESN'T DEPEND on anthropic or openai. their own model (composer 2) competes directly! yesterday sam altman said the ai model and harness are one and the same and today cursor turned their harness into a self-owned moat fucking masterclass (coming from a former cursor / ai wrapper hater)