Vishal

Software engineer | Building EnterviewGo

[Vishal @hellovisal]

How do you know that this idea will work?

[Vishal @hellovisal]

@rajputwt Congrats on the intern role. Landing an AI engineering role right out of the gate says a lot about the work you put in before this

[Vishal @hellovisal]

@qasimbizs Sitting on an idea for a year before shipping is more common than people admit. Curious what finally made you pull the trigger and launch it

[Vishal @hellovisal]

@anvithbuilds 1500 people choosing to learn alongside you is a solid signal youre sharing the right stuff. Congrats on the milestone

[Vishal @hellovisal]

@yevloop This is such a clean way to put it. A tutorial is basically an admission that the interface didnt do its job

[Vishal @hellovisal]

@dark_coderz Agreed. Vibe coding gets you to a working demo fast but the moment you need proper error handling, auth and scale, you end up rewriting most of it anyway

[Vishal @hellovisal]

@itssubhamoy No DSA and full agentic AI focus feels like where interviews are actually heading now. Good sign you were tested on real tool usage instead of whiteboard tricks

[Vishal @hellovisal]

@purvaasays Next step is a performance review where the agent explains why it missed the sprint goal

[Vishal @hellovisal]

@tiwarisuhani_11 This is the mental shift that takes a while to click. Indexes feel like a free win until you watch insert heavy tables slow down because of them.

[Vishal @hellovisal]

@sickdotdev The real skill shift is going from writing code to reviewing code you didnt write, which is a completely different muscle and nobody trained us for it

[Vishal @hellovisal]

@riteshpatel1884 The job description asking for 5 years of a framework that came out 2 years ago is the real punchline here

[Vishal @hellovisal]

@himanshuxspark The orchestrator framing is solid, thats basically the direction the whole industry is moving. For a first article the structure holds together well, maybe tighten the opening so it gets to the actual argument faster

[Vishal @hellovisal]

@BratDotAI Yes, but only because the alternative is worse. A flat sub means I either overpay on light months or ration my usage, pay per use at least matches cost to how much the model actually helped that week

[Vishal @hellovisal]

@Sheetal2205 Most places just want to see you can explain the tradeoffs, not implement RAG from scratch on a whiteboard. They ask coding when they want to check if you actually understand what a library like langchain is doing under the hood

[Vishal @hellovisal]

@nickkirt Depends on whether anyone outside your own head knows the thing exists yet. If nobody does, marketing this week just means shouting into a void, build first

[Vishal @hellovisal]

@Rina3AP Congrats. Senior isnt about having fewer unknowns, its about being calm enough to sit with them while you figure it out anyway

[Vishal @hellovisal]

@daryachach The marketing after launch line is the real lesson here, most people stop right at shipping and wonder why nobody shows up. Good luck with 10Goals

[Vishal @hellovisal]

@boringdev77 This is why I write the reasoning as a comment right above the weird code the moment I ship it. Future me at week 3 has zero memory of what past me was thinking and treats old context like archaeology

[Vishal @hellovisal]

@riyasayshie Consistency beats intensity here. Even 30 minutes a day on DSA compounds way more than a burnout weekend once a month, and it leaves you enough energy left for internships and college too

[Vishal @hellovisal]

@anvithbuilds Splitting the pipeline that way makes sense, each model is doing the one thing it's actually good at instead of asking one tool to do pose, motion and compositing all at once. How long is the full render for one clip end to end

[Vishal @hellovisal]

@dark_coderz System design is what determines if the thing still works when ten times the load hits it. Coding just gets you to a working demo, design is what keeps it standing after