sun zmey

@romxdev Do you type out your html tags by hand? Maybe someday someone will walk up behind your desk and see you hand typing code in notepad and think “wow you are so smart”.

Hot take: LLMs did not make bad developers better, they just allowed them to produce terrible, unmaintainable code at a much faster rate.

@usr_bin_roygbiv You can learn all of that in an hour.

you should know data structures, algorithm design, how a database works, C, compilers, assembly, git, unix based OS stuff and filesystems 100% and if they aren't teaching it drop out and stop wasting your time/money. Get an IT job or something around actual engineers

At the core entry developers are unable to work with code not generated by themselves, AI will seem foreign to them and untamed. A dev that is experienced with working on teams or unfamiliar environments will easily take up on AI coding (Interviewer thinking: not the answer I was told to get, but damn this sounds better).

Interviewer: Everyone has the same AI. So why are some devs shipping great products and others just shipping slop?

@404not_utkarsh AI is the only thing keeping this economy afloat, the same AI that was driving your archenemy in video games 25 years ago, the one still running on NVIDIA. The fact is you are wrong, and being sarcastic. This is because you already know you lost, your post sounds like a tantrum.

Folks who proudly say: "I am a vibe coder", "There is no need to learn coding", "Frontend is dead", "I can build it in 2 hours🤡" the slop they produce :

AI will be local in the near future, and at $5000-$10,000 upfront costs it will be similar to buying a house 50 years ago, you could still afford to buy it with your paychecks. Anyone who delays investing into memory chips, graphic cards, etc will be slowly getting left behind as entry price will only increase. Someone running a local llm in 10 years will have viable options on how to generate an income, the hosting environment will become the residence.

vibe coding is fun until you realize you’re just retrying prompts instead of actually learning how things work. if AI disappeared tomorrow, would you still know how to build it or debug it?

@Ric_RTP Its a higher cost, higher performance solution. Running cheap humans (usually Indian) does not mean better results.

Microsoft just banned its own engineers from using AI. The tool was literally costing MORE than the humans it was supposed to replace. They lied to you about AI adoption and now the whole narrative is blowing up: Microsoft gave thousands of engineers access to Claude Code six months ago and encouraged them to use it. Engineers loved it and adoption exploded. But then the invoices arrived. Token-based pricing means every query, every code review, every debugging session costs money. At scale across 100,000 engineers, the numbers became so large that Microsoft issued an internal order to cancel nearly all Claude Code licenses by end of June and force everyone onto their own cheaper tool instead. The company that invested $5 billion in Anthropic just told its own people to stop using Anthropic's product because it costs too much. Uber's story is even worse... Their CTO Praveen Neppalli Naga told The Information that the budget he planned for the full year was "blown away already" by April. Uber had rolled out Claude Code in December 2025. By March, 84% of their 5,000 engineers were using it with 70% of all committed code coming from AI systems. Heavy users were burning $500 to $2,000 per month each. Naga himself spent $1,200 in a single two-hour demo session. The company had even built internal leaderboards ranking engineers by how much AI they used. They literally gamified the spending and then ran out of money. Now look at what Nvidia's own VP of applied deep learning Bryan Catanzaro said to Axios last month. Direct quote: "For my team, the cost of compute is far beyond the costs of the employees." This is a VP at the company that SELLS the chips saying that using AI is more expensive than paying humans. Think about what this means for the entire AI narrative. Every CEO on every earnings call for the past two years has said the same thing: AI will make us more efficient, reduce headcount, and cut costs. The stock market rewarded every company that said it. Fired workers, stock goes up. Announced AI adoption, stock goes up. But the actual companies deploying AI at scale are discovering the math doesn't work. The MORE employees use AI, the HIGHER the bill. Goldman Sachs forecasts a 24x increase in token consumption by 2030 as companies adopt AI agents. Gartner just published a report showing that even though individual token prices will drop 90% by 2030, total enterprise AI costs will go UP because agents consume exponentially more tokens per task than basic tools. Meta built an internal dashboard called "Claudeonomics" to track which employees use the most AI. Amazon started pushing engineers to "tokenmaxx," their internal term for consuming as many AI tokens as possible. Both companies are spending hundreds of billions on AI infrastructure this year alone. And Microsoft, the company that bet its entire future on AI, just told 100,000 engineers to stop using the tool they liked best because the per-token bills got out of control. The companies building AI are telling investors it saves money. The companies using AI are finding out it costs more than the humans it was supposed to replace. And even the company that makes the chips just admitted it through its own VP. This is the gap nobody on Wall Street is pricing in. $725 billion in AI infrastructure spending this year across Big Tech. And the first companies to actually deploy these tools at scale are already pulling back because the economics don't work. What do you think?

@javarevisited I guess hb1 hires are done

If AI replaces 80% of coding work, who survives in tech careers?

@ThePrimeagen Your job turned from writing 1000 lines of code to writing 100,000 lines (period). The client now expects you to build entire systems instead of just a working script, in half the time, for less money. Knowing code allows you to orchestrate at this speed (period).

I am remembering right now for my own sanity. There is so much peace in deep thought instead of some insane frenetic agent orchestration fever dream

@asaio87 Writing code by hand means a disability of working with code not written by you. No ability to understand that the ai coder might not deliver if you do not provide clear instructions. Entire (deadweight) companies still hold on to this culture.

You want to say there are companies out there that have devs who write 90% of their code by hand ? Still ? I understand manual code review But write code by hand in 2026? Dude…

@WhaleNoName It doesn’t work that way, you are making the overlap happen based on an assumption. Instead you need to look at macro context, what is actually driving the market? You can output similar charts, but the catalyst is different. Today’s market is nothing like the dotcom era.

🚨 I CANNOT LOOK AWAY FROM THIS CHART 🚨 This is the Dotcom crash overlaid on today's S&P 500 Every week I come back expecting it to finally diverge It doesn't The overlap keeps tracking with a precision that is honestly unsettling What are your thoughts on this?

@Yuchenj_UW I still try to add up numbers in my head instead of using a calculator to show everyone how smart I am

Can’t believe I coded by hand for 15 years. 15 years of memorizing syntax, Vim, Stack Overflow, broken builds, cursed dependencies, merge conflicts, and “one last bug before sleep.” All of that just to end up typing “fix this” into a chat box and watching an agent do crimes.

@IamAroke Python/Javascript, all you will ever need

Javascript, typescript and python is slowly turning into vibe coders programming language 🤔

@_adityaa21 This is what all the diversity/feminism hires are going through, it sucks getting replaced by an AI

Tech industry right now is honestly insane. People outside think software engineers are just sitting in AC rooms earning easy money. Reality: - Developers are working full-time jobs - Grinding LeetCode after office - Learning AI tools at night - Building side projects on weekends - Upskilling constantly because tech changes every few months - Competing with people from the entire world And even after doing all this… one bad interview can make someone question their entire career. The pressure in tech is becoming unreal. You are expected to: - Be good at DSA - Know system design - Understand AI - Build projects - Have experience - Be active on LinkedIn - Stay updated daily And somehow still maintain health, relationships and mental peace. A lot of developers are not lazy anymore. They are simply exhausted.

@Sgtb45Hdhdhdhdh @0xglitchbyte I guess bottom of the barrel coder knows everything better (not)

@maxfilev @0xglitchbyte This couldn't be further from the truth. Sounds like cope from a vibe coder

If youre still writing code youre unironically so far ahead dont stop

@_adityaa21 In that case start your own path

One of my online friends shared this with me yesterday. Working full-time job. Sleeping 4 hours daily for months. No gym. No social life. Just office + LeetCode + stress. He solved graphs, DP, trees… grinded every single day. Then in the interview they asked some random O(1) data structure problem he had never seen before. Rejected again. And the worst part? He genuinely loves tech. Distributed systems, databases, backend engineering… that’s what excites him. But now he feels like he’s not smart enough because interviews keep breaking him. Sad reality: A lot of talented developers are burning out trying to crack interviews instead of actually becoming better engineers. One bad interview doesn’t mean you’re a bad engineer. But this industry really makes people feel that way sometimes.

@MegaBasedChad Because you know you you still program like a 12 year old (likely php or something) and your days are numbered

Excitement for AI programming seems to linearly decrease with programmer experience

@valigo Have it all outlined in a .md base before any vibecoding, update .md base after each build. I have a custom skin for mine with a few extras I added.

Vibecoding is when you let LLMs generate code and you accept it as long as the end result appears to be working. You don't even review the code, you don't care. When you rigorously ensure that generated code is up to a standard, it's not vibecoding anymore. It's programming.

@fjzeit Composer 2.5?

you'll all be lode coding or something similar within 6 months...

@i2cjak Login to your machine from your phone using Google Remote Desktop, works great. You can even vibe code while driving (hands free of course).

software “engineers” are so retarded. TAILSCALE. VPS. DISABLE SLEEP WHEN YOU SHUT THE LID. what is wrong with these idiots a fucking ELECTRICAL ENGINEER has to explain this to you