Ask HN: How do I get out of kdb+/q (and other niches)

[kdbQuestion]

Hi HN.

I have worked as a kdb+/q developer for 3 years now. I enjoy the language and the money is good, but I hate working in finance.

I was wondering if anyone had any experience getting out of an odd niche into general SWE?

Maybe you were a COBOL / Fortran / Common Lisp dev and you managed to transition to Python / Java.

Any and all advice would be much appreciated!

[secondaryacct]

Super interesting, I work in finance and love it, but I loathe kdb: it's impossible to read your own code 2 weeks later, every function is one letter, so go google what -9!' does exactly, quants work in isolation rather than in teams so code sharing is done via mounting the prod shared drive with all the prod code and so on.

I d suggest you do the contrary of me (who dabbles in KDB to help the Quants out of their mess) and try to join / work on more java/C++ features where you are so you can gain xp.

[giantg2]

I worked on Filenet and Neoxam based systems. Try to get some certs, like AWS. See if there is any common stuff between your current job and desired job you can learn, like scripting.

Sometimes it can be easier to switch at the same company. See if you can get into a different team that works with the newer tech. If you are switching companies you may need to take a pay and/or title cut.

I work in finance too and hate it. I'm not sure anywhere else would be better though.

[bradwood → giantg2]

> I work in finance too and hate it. I'm not sure anywhere else would be better though.

There are definitely better options than finance. I worked in finance (investment banking) for years. Didn't like it either.

The obvious next move is to a small/medium size fintech shop. Ideally one not encumbered by regulatory straight jackets and that is making a cool product. These can be more agile and responsive to customer needs and a lot of fun to work at.

[muzani → bradwood]

I work in finance and I absolutely love it. We're pretty big. Half the work is regulation, especially in places where corruption and crime is rampant. But we still get to make cool products.

I think the trick is cutting down overhead because there's already so much friction inherent as it is. Management skill is essential, as every hour management buys for developers gives back good returns. Something like defensive programming is also a great return on investment as bugs just leads to more processes and chaos later.

[bradwood → muzani]

Whilst management often sucks in big banks. What gets me down is the process drag. Everything is ITIL signoffs, death by bikeshedding, and quarterly releases. Silos abound and the pernicious affect of Conway's law applies to turn most of it into a massive cluster fuck of spaghetti and string.

No thanks.

[giantg2 → bradwood]

I actually want more structured work. I work for a large company. They are being too "agile" (a poor implementation of Agile) and there is tons of context switching. I already feel too old to switch. I've basically given up and realized I have a job, not a career. I have no motivation.

[Pet_Ant → giantg2]

Welcome to the industry. That's just what it's like. The code is trivial, getting clear direction is the challenge. Just cash the cheques and get some hobbies.

[giantg2 → Pet_Ant]

The checks aren't very good and it leaves little time for hobbies.

[zaphar → giantg2]

Start interviewing. There are places out there where the pay is really good and you have time for hobbies. They may still have similar pathologies as other places but at least you'll have a life outside of work. You don't have to be stuck at a place that sucks everything out of you and gives nothing back.

[giantg2 → zaphar]

I've been looking, but not finding anything better. I've basically wasted the first 10 years of my "career".

[readonthegoapp]

I would guess it would take you upwards of 24 hours to get out if you wanted to.

[eismcc]

I’ve spent a fair amount of time recently reading about kdb (and family) and I’d think a lot of the skills you need to use it are very applicable to other systems. You may even be further ahead of the game than you realize. The main gap I’d guess would be the jump from finance process to product process. That’s just time and I’d just start interviewing.

[jshaqaw]

KDB is trying to expand out of the finance niche into other applications no? Is that a viable path? Is KDB competitive in other sectors?

[randomopining]

How do you get into it? And how much does a mid-level kdb dev make?