Console.log(me)

I don't care about being a "content creator." I care about shipping code that solves actual problems. Less tweeting. More building. (Updates only when I have something worth showing.)

@TechByTaraa None of these, all these give you headache in the long run. Prefer codex or just Byok with cheaper opensource model in vscode.

Hey devs, which is the best vibe coding tool ?

@msdev The usage limit is a joke though, even use it conservatively can cause more than codex

GitHub Copilot CLI is even more powerful when you know the right commands. Save this cheat sheet to speed up your workflows from the terminal.

@ritu_twts It doeant matter with other choice but stay away from Antigravity because it is super buggy and have usage limit bug, and copilot is getting so expensive soon that you better off with codex or just BYOK with cheaper alternatives

I’m ready to invest $20, which one should I choose? - Claude - Codex - Cursor - Antigravity - GitHub Copilot Which one is more worth it right now?

@sinclairinat0r Yeah their pricing is absurd now, I am super conservative and never run anything in parallel also can hit such high charges, time to change

Lmaooo yoooo .... Wordddd???? This is a Copilot Billing Preview for the usage-based billing. I hope this estimate is $2646.63 off from the actual price bc.....😂 I think we're going to start seeing the "Let's just write this in assembly to ultra optimize it to the tiniest size possible" being the move

@CodeWithAmann I am not sure if its a good idea but I'm keen to try alibaba cloud, not sure if anyone try it before.

As a developer, which one is the best hosting platform?

@zuess05 So true, some big company even have policies to install dev tool, mac, xcode, vscode upgrade is basically impossible, everything need approval even for simple tool chain or library upgrade 🤢

The gap between solo builders and corporate tech has never been funnier. One kid with a laptop and AI is shipping entire product ecosystems from their bedroom over a weekend. Meanwhile, Fortune 500 companies are still booking 45-minute Teams calls with 12 people just to get legal approval to change the padding on a button.

@Umesh__digital Drug dealer try selling drug, thats all

Can someone explain to me why Anthropic's CEO keeps saying Software Engineering is dead, yet his company is still hiring Software Engineers?

I consider myself to be very conservative with my usage on @GitHubCopilot , used not even half of my premium request, and always use surgical prompts for AI Assisted coding instead of vibe coding. Yet, the charges coming is so absurd. I can get a @OpenAICodexCli plan with this!!!

@zuess05 The actual case is juniors write 10x slops, and think it is senior level work.

I am actually curious. If junior devs are using Claude to write code at a senior level... And senior devs are using Claude to write code at a senior level... What exactly is the difference between the two roles right now, other than a $100k gap in salary?

@sukh_saroy That is exactly true, and it also end up saving a lot of trouble and wasted token in the long run, ultimately the cost. You get the speed, reliability, and cost efficiency equipping senior with AI. So yes, knowledge on coding and system design still matters

A new study just blew up the entire "vibe coding" movement. Researchers from UC San Diego and Cornell tracked 112 experienced software developers using AI agents in their actual jobs. The finding is the opposite of every viral demo on your timeline. Professional developers don't vibe code. They control. Here's what they actually found. The researchers ran two studies. 13 developers were observed live as they coded with agents in real production work. 99 more answered a deep qualitative survey. Every participant had at least 3 years of professional experience. Some had 25. The viral pitch of agentic coding goes like this. Hand the agent a vague prompt. Don't read the diff. Forget the code even exists. Trust the vibes. Andrej Karpathy coined the term. Tens of thousands of developers on X claim to run "dozens of agents at once" building entire production systems hands-off. The data says almost nobody serious actually works that way. Here is what experienced developers do instead. → They plan before they prompt. They write out the architecture, the constraints, and the edge cases first, then hand the agent a tightly scoped task. → They review every diff. Not because they're paranoid. Because they've seen what happens when you don't. → They constrain the agent's blast radius. Small, well-defined tasks only. The moment a problem touches multiple systems or has unclear requirements, they take over. → They treat the agent like a fast junior dev that needs supervision, not a senior engineer that can be trusted alone. The researchers also found something darker buried in the data. A separate randomized trial they cite showed that experienced open source maintainers were 19% slower when allowed to use AI. A different agentic system deployed in a real issue tracker had only 8% of its invocations result in a merged pull request. 92% failure rate in production. 19% productivity drop for senior devs. The viral demos lied to you. The paper's biggest insight is in one sentence: experienced developers feel positive about AI agents only when they remain in control. The moment they let go, quality collapses, and they know it. This matches what every serious shop has quietly figured out. The developers shipping the most with AI right now aren't the ones vibing. They're the ones with the strictest review processes, the tightest task scoping, and the clearest mental model of what the agent can and cannot do. Vibe coding makes for great Twitter videos. It does not make great software. The next time someone tells you they let Claude build their entire SaaS in a weekend, ask them how much of that code they've actually read. The honest answer separates real engineers from the demo crowd.

@ai_for_success If you're getting a dev machine, go with 128, you can run better model.

Need suggestions from people who use Mac Pro. 64GB or 128GB unified memory?

@adamemedia1 That's how it should be done, must always elevate what ai can do we more human input, not let AI do what it can without human adjustment. Efficiency matter more than every in the new age where energy and cost getting expensive and limited.

CHINA JUST DREW A LINE ON AI A court in China has ruled it ILLEGAL to replace human workers with AI purely to cut costs. They have put responsibility back on corporations. They can’t automate just to boost margins while workers are pushed out. China has decided that wages, fairness, and employment aren’t optional. And that’s a big shift. While the west races to replace labour as fast as possible, viewing AI as a free-for-all… China has set a precedent that profit alone isn’t enough and corporations must answer to society.

@zuess05 This is true, AI screw up very often and need engineer step in, while all projects manager do is repeatedly ask what is the progress. You can do it with AI.

The craziest realization of 2026: AI didn't automate the coding. It automated the meetings. While corporate teams spend 3 weeks in Jira planning how to build a feature, a solo dev just shipped it, posted a demo on X, and got their first 10 paying users. The ultimate moat right now is just not having a Scrum Master.

@immasiddx Gemini code assist is shit thought in vscode. I try switching to it from GitHub copilot, and Gemini assist is too slow to suggest, and worst of all, stuck while generating any Usable code in agentic mode. Antigravity too have model limit bug, and burned 1 week limit in 10 prompt

Google AI Pro is literally the most worth it AI subscription in the market right now. $20/month gets you: - Access to Gemini - 5TB cloud storage - Best in class image generation (Nano Banana) - Google Antigravity & so much more! No one can compete with this.

@ai_for_success Hands down antigravity, their usage is shit

Delete 1 from > Claude Code > Codex > Antigravity > Cursor

@jfversluis Byok model still available?

GitHub Copilot is moving to usage-based billing June 1. Premium requests → AI Credits (based on token usage). Plan prices stay the same. Code completions still included. Big change: no more fallback model when you run out. Credits gone = usage stops.

@vlad_mihalcea They still got weekly token limit though on top of premium request, need max out every week or will hit limit still

Starting in June, Copilot will use a new token multiplier model that will burn tokens rapidly on your annual subscriptions. If you have unfinished projects on your TODO list, you should take advantage of the current multipliers and use the month of May to do the AI Spring cleaning and finish everything while it's cheap to do so.

@edandersen Not just that, they say there will be "no update" for those still in annual plan, and that is bad

What’s incredible about this announcement is that annual Pro and Pro+ subscribers get the premium request model grandfathered in, but Opus 4.7 goes from 7.5x to 27x!!

@awakecoding So true. But I wonder if their Bring Your Own Model will still work after June 1 thought, if it still work, then there might still be workaround

I'm burning premium requests like there's no tomorrow until June 1st 😎

@github Shit, and I supported you for so many years, at least let existing annual plan subscriber to continue have updates and new model until the end of our plan. Now you charging us by marking up premium request and pull any update is unfair. I want refund at end of May.

Starting June 1st, GitHub Copilot will move to a usage-based billing model as GitHub Copilot supports more agentic and advanced workflows. In early May, you'll see a preview bill experience, giving visibility into projected costs before the transition. 👉 Read more about the upcoming change: github.blog/news-insights/…