Aryak Singh

@sreecharantwts Also the first contri is difficult but it give you so much confidence which no tutorial can give.

day 14/30 today's builder lesson. your first open source contribution isn't about writing great code. it's about learning how large teams communicate. reading discussions. accepting feedback. improving your work. the code is only half the contribution.

@realsohamparekh Non tech maghloos.

Recommend Wispr Flow to a friend. Got hit with, “I’m not menstruating.” Incredible product. Unfortunate name.

@sreecharantwts Congrats bro 🤝

just woke up to this

@GithubProjects github.com/aryaksinghh

_________________ ___________ | Share your GitHub profile | | ___________________________| ¯\_(•‿•)_/¯

@sreecharantwts Yeahh, sometimes models are too over advanced for a particular project which creates inefficiency

day 12/30 i think ai is entering a different phase. for the last two years everyone chased bigger models. now i'm seeing more excitement around smaller, faster, cheaper ones. turns out most products don't need the smartest model. they need the one users can actually afford.

Different priorities 😂

@kseniam0s covalentai.vercel.app (AI revision app for developers)

Hey founders, what are you building today? Let’s connect

@sreecharantwts This is out of my knowledge😂. Do you have a solution?

@aryakceo think of it this way: the payment provider says "payment successful." before your server finishes saving that information, it crashes. from the user's perspective, nothing happened, so they click "pay" again. how do you make sure they aren't charged twice?

backend interview question. your api creates an order. the payment succeeds. right before saving the order to the database, your server crashes. the customer refreshes and clicks "pay" again. how would you prevent charging them twice?

@sreecharantwts Thanks bro 🤝

@aryakceo really cool bro

Day 30 : Completed my ai revision app for developers. Live - covalentai.vercel.app Github - github.com/aryaksinghh/co… This is a revision app for developers with feynman learning methods. It has a learning engine full tuned to make your memory long term. Make sure to checkout.

@Vivek4real_ Just like google map 3d view?

THIS IS WHAT THE FUTURE OF REAL ESTATE LOOKS LIKE. POWERED BY AI. WILD TIMES AHEAD.

@itsstanyy Great bro

just got a ₹10k advance from a Netherlands client international client bhi mil gaya ab kaam bhi hai aur pressure bhi hai grinding time guys

@alwayspriyesh @X hey, currently i am building an AI revision app for developers. covalentai.vercel.app. lets connect 🚀

Looking to connect people on @X if you're into - design - building SaaS - vibe coding - AI tools - shipping in public - figuring it out as you go say Hi or drop what you're working on looking to follow active ones 👋

Day 30 : Completed my ai revision app for developers. Live - covalentai.vercel.app Github - github.com/aryaksinghh/co… This is a AI revision app for developers with feynman learning methods. Make sure to checkout and give star in github.

@gxjo_dev bro 12lakh/month ? Sounds like a joke 😂

JEE is the ultimate cheatcode

@rrhoover Humans can generate innovative ideas through their needs. But llm is just pattern matching no need feeling.

I don’t understand the argument that LLMs can’t come up with original ideas. “It’s just trained on existing data, predicting the next token.” How is that different than human ideation? We’re also pattern-matching machines, influenced by our life experience.

@shubhmx What is the accuracy rate of voice transcription?

introducing claude hands-free the fastest way to run your army of ai agents

@sreecharantwts So true. Code can be generated from LLM's but confidence not.

day 11/30 today's builder lesson. people don't pay for code. they pay for confidence. - confidence that it works. - confidence that it scales. - confidence that you'll still be there when something breaks. writing code is only part of the job. earning trust is the rest.

@sreecharantwts The biggest problem as a dev is over engeneering 🥲

day 10/30 today's builder lesson. i used to think writing code was the hard part. it wasn't. the hard part was: > deciding what not to build. > saying no to unnecessary features. > keeping the product simple. > resisting scope creep. > shipping before perfection. every feature has a cost. build carefully.