Ask HN: How did Soham Parekh get so many jobs?

> I can tell you it's because he's actually a very skilled engineer. He will blow the interviews completely out of the water. Easily top 1% or top 0.1% of candidates -- other startups will tell you this as well.

It is hilarious that companies that hired a guy who was scamming them are also convinced they are great at assessing the skill level of devs.

Being a good developer and being a scammer are completely uncorrelated variables.

Someone can be a good developer and also be a scammer. I don't understand why you think this is hilarious or weird.

Is it so hard to believe that someone can be a great candidate in an interview when you're getting 100% of their attention and then be horrible at their job when you're getting 20% of it because they're juggling 5 jobs?

It's hilarious because companies use such scammable ways to define who is "top 0.1%"

Also there's a ton amazing engs out there who want and need work but the companies all only want that one "perfect" guy (or gal), as if such a thing exists

> Being a good developer and being a scammer are completely uncorrelated variables.

One could expect good developers to be less inclined to fraud as they may not “need” it as much.

That also made me thing of Berkson’s paradox: https://en.m.wikipedia.org/wiki/Berkson%27s_paradox

If these were really independent traits they would look negatively correlated as we talk about people who are good OR scammers.

It's not about need, it's about beating the system. The "hack".

The “need” of beating the system. Good developers may or may not have a lower deficit of “it”.

Exactly. It's so bleak that this industry throws integrity out the window in the name of productivity.

IMO being a good corporate developer is not very correlated with being a good "hacker" (finding ways to exploit systems). They may be correlated a little but not very. Being a good startup founder is probably correlated with being a good hacker, much more than being a good corporate developer is. Startups have to find and exploit niches.

With due respect, they probably just asked leetcode-esque and sys design questions.

he had no proof he can code, no projects, no github, only hired because he gave them a lowball offer, it was lowball because he was scamming

The had 100 candidates and hired him. Top 1% QED. (/s)

There’s literally no evidence they did either of these things. I really hope these companies can explain their hiring process as it reflects badly on them that they keep calling him top 0.1% without any explanation of their process.

The OP said he blew interviews out of the water. Presumably they mean technical interviews, that's how he proved he can code. By writing code.

Lots of devs don't have personal projects. I love programming but after spending the whole day programming I don't particularly want to go home and continue programming.

I've seen a lot more employed shitty devs than I've seen unemployed amazing devs. In fact I don't know a single competent developer who has trouble getting work.

This is in Norway, maybe it's different elsewhere.

My observation of the situation in Germany is different: I know quite some quite skilled developers who have/had quite some trouble getting jobs.

The other hand, the bad developers (impostors) who easily found jobs were typically sycophants. On the other hand, many highly skilled developers were rather more stubborn (they can't stand bullshit), very honest, and not "corporate politicians".

Where did you hear this? People on X said the exact opposite.

There’s low correlation between success and ability once you hit a baseline.

There’s an element of chance, and stuff like leetcode is just a veneer of science over a vibe based process, which conveniently scopes out management culpability. Personally, I think it’s hilarious that Silicon Valley types have essentially enshrined civil service exams for hiring.

you wouldn't and he wouldn't get past me, lots of devs don't have personal projects? That is a good filter, give me the ones with higher agency that do. Maybe my standards are too high. I usually contribute the most lines at work, and have personal projects that helped get me the job in the first place.

With agentic coding on the rise, it's easier to make a side project after a year of nights and weekends.

Nobody wants to go home and make a project that proves to future employers you can do the job, everybody wants to get paid though.

Technical interviews and answering system design questions do not prove you can code. This is for a founding engineer position. When I started in the industry in 2014, I had to explain the architecture decisions of my side projects. Now people parrot information off the web and call it proof of code.

I am like you, I'm someone that can happily code for eight straight hours on task and then happily jump to a side project afterwards in a totally different language / domain. Take it for the single point of data that it is, but: I have met equally as many high performers who do NOT code at home as I have mediocre ones that did. And the number of programmers who code solely on the job is far, far more common than those who take their work home with them. Statistically speaking, at least in my personal experience, that makes the at-home coder notably worse on average. As someone that used to think like you, I think the only thing you're really doing is the boys-club-esque equivalent of what the old Ivy League managers used to do (Oh, he's from MIT! That's perfect, we only hire MIT grads here). You're hiring what you know because you understand people with that background more.

You might find on the opposite end of the spectrum that someone with a perfectly equal skillset is laughing at the idea of taking on personal projects when he can just optimize for getting hired somewhere where they prioritize paying you to learn on the job.