Prakhar Yadav

0 → 200. 200 strangers decided my mess was worth following. I came here with: — Zero content plan — Zero audience — Zero confidence Just a 4th year Tier 3 student with 3 years of wasted time and too much honesty. 200 people said "I want to watch this." Not because I'm successful. Because I'm not. And somehow that's more relatable than any success story. Thank you. Not done. Just getting started. Next stop: 500. Same honesty. Same mess. More building.

@HxrshitYadav This is honestly very real. The market is tough, but I like that you’re actually doing something instead of just waiting. Building real projects and talking to people directly probably matters more than ever now.

I'm graduating in 2026 and honestly the timing couldn't be worse. > Entry level jobs are disappearing. > AI is eating the basic work freshers used to get hired for. > Companies are posting fewer roles and getting 500 applications within 2 hours. > The 2022 batch had it on easy mode compared to what we're walking into. So here's what I'm actually doing about it. Not advice. Just what I'm trying: > Stopped building tutorial projects. Every fresher has a todo app and a weather app. I have them too. They've done nothing for me. Building something that actually solves a problem now even if it's small and ugly. > Stopped applying cold to job portals and waiting. Doesn't work anymore. Reaching out to actual people who work at places I want to join. Not "Bhaiya refer krdo" Actually talking to them. Showing them what I'm building. > Stopped waiting for the perfect opportunity. If a startup offers something small I'm taking it. Any real work experience right now changes everything about how the next application looks. > Picked one stack and going deep. Java and Spring Boot. Not 5 languages at surface level. One thing properly. > Stopped waiting to feel ready. Nobody gets placed when they feel ready. They get placed when they stop waiting. I don't know if any of this will work. I genuinely don't. But sitting and hoping the market gets better is not a plan. If you're also graduating in 2026 what are you actually doing differently? Not what you plan to do. What you're doing right now.

Amazon India sent me an Evofox wired keyboard when I ordered a Kreo Swarm 75 Wireless. Box says Kreo Swarm. Product inside? Completely different brand. @amazonIN @AmazonHelp what exactly are you checking before shipping? Sort this out.

@brankopetric00 That’s a classic oversight. Rate limiting by IP works fine until you remember that offices, campuses, or even mobile networks often share the same public IP. Suddenly hundreds of normal users look like one aggressive client.

Rate limit was set per IP address. Entire office shared one IP. 200 employees hit rate limit in 10 minutes.

This is actually a really balanced take. WFH can save a lot of money and give you a better lifestyle in many ways. But sometimes people forget that careers also grow through visibility, conversations, and being around the right people. It’s not that one choice is right or wrong. it’s just a trade-off. You save on cost of living, but you might lose some of the networking and exposure that naturally happens when you’re closer to the action.

My friend chose WFH and decided to stay in his village. Salary -> 50 LPA Cost of living -> almost 0 LPA Everyone said he made the smartest decision. Bangalore rent saved -> 2.5 LPA Travel cost saved->1LPA Home food every day. On paper, he was saving lakhs every month. But reality was different. Tech meetups in Bangalore -> missed Manager visibility -> low Team bonding -> almost none There was a startup meet in Banglore, but he couldn’t go. Meanwhile village life had its own surprises. Relatives dropping in anytime -> daily interruptions Local functions -> “you must attend” Random responsibilities -> unavoidable Quiet focus time -> rare Slowly something changed. He was still working. But he wasn’t really in the loop anymore. Decisions happened without him. Opportunities passed quietly. His network stopped growing. Money saved. But career momentum slowed. Sometimes the most expensive thing isn’t rent. It’s distance from the room where things happen.

I kind of relate to this. AI didn’t really ruin coding for me either. If anything, it removed a lot of the repetitive stuff that used to take hours. The interesting part was never typing boilerplate anyway. it was figuring out the problem and designing the solution. The real fun in programming is still there. The tools just changed.

"AI ruined coding" is the hottest take in tech right now. But what exactly did it ruin? Writing the same CRUD endpoints for the 400th time? Debugging webpack configs? Copy-pasting from Stack Overflow? AI took away the parts I hated and left me with the parts I love: architecture, product thinking, creative problem solving. Coding hasn't been this fun for me in 7 years.

It’s honestly one of the craziest stories in tech. A student builds something as a personal project, shares it on the internet, and decades later it ends up running most of the world’s infrastructure. That shows how powerful open source and curiosity-driven projects can be. Also a good reminder that a lot of the biggest things in tech didn’t start as companies — they started as someone just building something interesting.

Linus Torvalds created Linux at 21 without Claude or any other AI. - He didn't have a co-founder. - No VC funding. No office. - No team. - Just a personal project he posted to a mailing list: "I'm doing a free OS." 33 years later, it runs 97% of the world's servers, all smartphones, and the International Space Station. The most important software in history started as someone's side project. Absolute legend.

Yeah, I’ve noticed that too. The models themselves are really impressive, but sometimes the apps around them feel surprisingly rough. You’d expect companies with that much funding to put the same level of polish into the UI and performance as they do into the models. Tools like Codex feel smoother because the experience around the model actually matters just as much as the model itself.

Claude’s desktop app is a joke. UI, UX, performance - everything about it is bad. I don’t understand how a company with infinite money can’t hire someone, or at least allocate resources, to fix it. Compared to it, the Codex app is a masterpiece.

Pattern recognition is definitely a big part of intelligence. But I think it’s only one piece of the puzzle. Real intelligence also shows up in things like judgment, creativity, asking the right questions, and understanding context. Recognizing patterns helps you see what’s happening — but deciding what to do about it is a different skill.

pattern recognition is the highest form of intelligence

From what I understand, the difference is mostly about context and role. An assassin is someone specifically hired to secretly kill a particular person, usually a high-profile target. The focus is on the act of assassination itself. A hitman is basically a contract killer, often associated with organized crime. They’re hired to eliminate someone for money, but usually in a criminal underworld context. A mercenary is different. They’re a soldier who fights for whoever pays them, usually in wars or conflicts. Their job isn’t necessarily assassination. it’s more like combat or security work.

can someone tell me what the difference between an assassin, hitman, and a mercenary is

This is a really good point. AI can help you get a prototype up quickly, but building something that actually survives real users is a completely different game. Things like concurrency issues, edge cases, data integrity, and system design don’t magically disappear just because the code was generated faster. AI is great for speeding things up, but someone still needs the experience to spot what could break before it hits production. The hard part of software has always been understanding the system, not just writing the first version of it.

Someone builds a project management tool with Claude Code over a weekend. Ships it. Tweets "just replaced Jira." The app works. One user, happy path, localhost. Then two people edit the same record simultaneously, and the data is silently corrupted. They don't know what an optimistic lock is. They never needed to before. The prototype is maybe 1% of what makes software actually work. The other 99% is what you find after real users show up: race conditions, failed transactions, sessions expiring at the wrong moment, a payment webhook that fires twice and charges someone double. AI didn't cover any of that. It built exactly what you asked for. And the confidence is the worst part. "Just need to adjust a few things before we go live." The few things you need to adjust are the product. That's like laying a foundation and telling people you basically built the house. Vibe coding works. For personal tools, throwaway scripts, and prototypes you'll never put in front of paying users, it's genuinely fast and good enough. I use it. But there's a hard ceiling, and it shows up the moment the stakes get real. Agentic engineering is a different discipline. You're not prompting for code. You're decomposing problems, designing system boundaries, writing specs precise enough that the agent doesn't go sideways. You review everything it builds, because it will make mistakes that only look wrong if you know what correct looks like. You guide it. You catch what it misses. If you don't know what a distributed transaction is, the agent won't save you. It'll generate something broken with complete confidence, and you won't know until production. The hard part of software was never writing the first 200 lines. It never was.

I don’t think software was eaten by AI. If anything, AI is just changing how software gets built. The tools are getting smarter, but someone still has to understand the problem, design the system, and decide what should actually be built. The code might come faster now, but the thinking behind it still matters a lot.

Software was eaten by AI.

Honestly… this is true. Bad code didn’t suddenly appear because of AI. Developers have been writing messy, rushed, or hard-to-maintain code for years. AI just makes it easier to generate it faster. The real difference is still the same as before someone has to review it, understand it, and clean it up.

Devs are acting like they didn’t write slop code before AI.

@billcompute Had me in the first half 😅 For a second I was imagining a full data center of Mac Minis running agents. The internet really does make everything sound believable until the last line.

I currently run 412 OpenClaw agents 24/7 across 31 Mac Minis. They’re: • building apps • automating my life • generating business ideas People ask how to replicate this system. It’s actually very simple. You just need about 30 Mac Minis, a solid imagination, and absolutely no fear of making things up on the internet.

@TicTocTick That’s actually one of the few subscriptions that feels worth it. You hope you never need it, but the moment something breaks, it suddenly makes sense. Getting a replacement that quickly without a long process is honestly pretty solid service.

I pay 39 dollars a month for apple Care (insure 3 devices ) . I never use case on any of my iPhone . Yesterday I dropped my iPhone and broke it . Apple sent me a replacement phone today, no questions asked . For $99 one time payment . Great company.

Honestly, I’d probably choose Option B. Money and freedom sound great, but a loud house, kids running around, and a full dinner table feels like a different kind of wealth. The chaos might be exhausting sometimes, but it also means life is full. In the long run, I think those moments matter more than a bigger bank balance.

Imagine you had to choose your life at age 40: Option A: Single. No kids. $10M net worth. Travel anywhere. Total freedom. Quiet house. Quiet holidays. Option B: Married. 3 kids. $1M net worth. Drive a Toyota. Chaos every morning. Loud house. Full dinner table. Be honest, which life are you choosing?

Interviewer: If passwords in a system are stored as hashes and the original password is never saved. how does the “Forgot Password” feature work?

Yeah, when you really zoom out it’s kind of mind-blowing. At the lowest level it’s literally just tiny switches turning on and off, electricity moving through silicon. But when you stack layers on top of that logic gates, assembly, programming languages, operating systems suddenly those little signals become apps, games, AI models, entire digital worlds. It’s wild that something so physical and simple underneath can produce something so complex on the surface.

dude computers are actually so fucking insane when you really think about it. we literally figured out how to write some fake-ass rules called code and somehow convinced rocks to follow them. like actual rocks. sand, melted, purified, carved into tiny pathways where electricity just flows in patterns. that’s it. that’s the whole magic. and yet from that we get operating systems, compilers, kernels, networks, distributed systems, machine learning models, entire virtual worlds running inside other virtual worlds. billions of tiny electrical decisions per second, all because we defined some abstract logic. humans basically invented a language of instructions and taught matter itself to execute it.

Honestly, I’d probably spend it on things that buy time and freedom. I’d invest most of it index funds, some solid assets so it keeps growing instead of disappearing. Then I’d spend a part on experiences: travel, learning new skills, maybe funding a few projects or startups I actually care about. Because $2M on random stuff disappears fast. $2M used well can change the rest of your life.

You’re given $2m. You have 20 minutes to spend it. You can’t spend it on cars, airplanes, yacht or a house. You can’t spend it on golds or diamonds either. What will you buy??

This is actually the part people don’t talk about enough. When someone hears ₹1.5L salary, it sounds huge back home. But once you factor in rent, transport, food, and just the general cost of living in a big city, the number starts shrinking pretty fast. That’s why salary alone doesn’t tell the full story. Cost of living, savings, and quality of life matter just as much as the paycheck.

My cousin moved to Bangalore for a ₹1.5L/month tech job. Six months later, he realised something no one tells freshers. Computer Science graduate First software developer job Moved to Bengaluru for better opportunities Salary: ₹1.5L per month Back home, everyone thought he had made it. Big tech job. Big city. Big salary. But a few months later, reality looked different. Rent for a small apartment near the office: ₹36K Food & groceries: ₹13K-15K Cabs & autos (traffic is insane): ₹6K-8K Swiggy, coffee, weekends out: ₹10K-12K Then come the things nobody talks about. Subscriptions. Medical expenses. Unexpected bills. Sending money back home. By the end of the month… Savings were barely ₹15K-20K. That’s when he realised something most people outside big cities don’t see. ₹1L+ salary sounds huge in your hometown. But in a city like Bangalore, it often just means you’re getting started. Big salary. Small savings. Welcome to Banglore life.

That’s actually a great catch. A lot of people don’t realize that giving the model structured context like state.md or architecture.md can make a huge difference. Otherwise it just scans everything and wastes tokens figuring things out. Small setup, but massive improvement in speed and efficiency.

okay why did nobody tell me about state.md for claude code? i've been stuck wondering why a project i build on daily was taking 8-10 mins to make changes turns out claude was reading EVERYTHING (all my code) before it made a change. i added state.md, architecture.md (alongside my claude.md) and bam, 8,000 tokens to ~1,000 tokens and from 8 mins to ~ 60 seconds. claude.md wasn't enough in this instance annnnd i'm so happy lol