Hacker News Reader

Comment by game_the0ry | Hacker News Reader

I worked in a brand-name bank, so I am familiar with the goings in the tech side of finance.

Call me crazy, but I would trust aws, microsoft, and google with my PII and finances before I would trust Wells, BofA, JPMC, Goldman, et al.

The cloud giants pay their engineers more and technologists are second-class citizens at the financial institutions - infer what you want from that.

View on HN

Definitely agree. Having worked extensively with some large banks I've found that they are absolutely cutting edge in technology in one area and one area only: credit card fraud analytics. That's because credit card fraud creates a material financial risk for the bank that they can't defer to someone else. Most other risks banks deal with are managed through complex financial hedging, government backstops, or insurance schemes and so they invest very little in doing anything "correctly". Rather, almost everything about banking compliance is driven by checkbox tickers and not by experts.

I think it is unfair to think that banks don't want to do things correctly but they suffer along with all other corporates of too many levels of management to be agile, systems that are far to risky to change/replace, legacy systems and a high turnover of staff. If I ran a bank, I suspect I would run it exactly the same way because I have to.

Perhaps I'm too close to the issue as I have worked with these businesses, but I disagree. This is intentional on their part, because banks first and foremost are about identifying and managing risk. If they can find a way to mitigate that risk or externalize it, then they no longer need to deal with it head-on. Most of the risks related to IT should be dealt with head-on because they grow over time as you stay still and technology moves further away from you, and it's impossible for anyone to accurately predict the future and therefore future risks.

Many banks and other financial institutions are finding this out now in their desperate bid to find competent COBOL programmers to continuing maintaining legacy critical applications running on mainframes when most COBOL programmers are retired or dead, and more are headed that way every day. It wouldn't shock me to find out that the effects of COVID being weighted towards worse outcomes for older people had a material effect on the human resource risk of using COBOL-based critical systems.

Banks will absolutely hold on to anything to avoid an unknown risk as long as they think they can hedge or mitigate known risks, and utterly fail to acknowledge the truth of unknown unknowns. This will ultimately be their downfall if governments ever let them follow standard business outcomes, otherwise they'll eventually absorb some upstart to keep hedging forever.

>Banks will absolutely hold on to anything to avoid an unknown risk as long as they think they can hedge or mitigate known risks, and utterly fail to acknowledge the truth of unknown unknowns

Nassim Taleb has written entire books on this fact about banks. It's interesting that the same flaws in their financial thinking apply to their IT infrastructure.

I think technology is fundamentally a tool for organizations, and as such will always evolve to reflect the larger organization. You can see this in software design and it is also seen in the IT infrastructure architecture.

My understanding is that at least Chase, Blackrock, and Visa are rapidly raising their pay for new engineering / analyst hires. Is this not true?

I can speak for the bank I worked at - this is true, but the base pay level was very low to begin with and the increase is nowhere near competitive for the most talented engineers. Also, the experienced engineers at those companies are salty about new hires making more then they are.

Again, this stems from the culture of financial institutions where tech workers are viewed as support staff, not strategically critical, despite what financial institutions say publicly.

I know for a fact that base pay for a first year SWE with a BS+MS in CS at Goldman just broke $100k a couple years ago. That probably sounds like a lot to most people, but if you're hiring staff to live and work in NYC, especially when there's an absolutely massive Google campus 10 minutes North of the GS HQ, that's pretty much table stakes. Never mind the glaring cultural differences.

Does that take bonuses into account? Goldman (and others in the industry) tend to have lower base pay with a much higher bonus. It's typical for low level employees to get a year-end bonus equal to about 50-60% of their salary. As they advance to mid-level their bonus will equal their salary and, if they're in a senior position, it will greatly exceed their salary.

Bonuses are not that high at Goldman and their comp still does not compare to FAANG, it never will.

Keep in mind it easily could at the stroke of a pen though. It's entirely their choice to not compensate competitively.

https://tipalti.com/profit-per-employee/

Visa regularly makes double the profit per employee of FAANG - so, yes, they could easily pay more.

Thanks for the source to backup that feeling.

Don't all the smart companies avoid making any profit at all, since profit is taxed (but share buybacks are not)?

You have to buy your shares with FCF (usually profit) or debt.

They have profits. They just aren't in the US.

"Back office" (which includes most programmers, with the exception of algo traders) bonuses are much much lower than "front office" (traders, ibankers, sales) bonuses.

Google/FB devs total comp (including RSUs) will easily beat Goldman devs by 150-200%, depending upon your performance.

This is strictly base pay, but the bonuses are simply not that good for the average SWE at the banks, because as others have said, technologists are seen as a supporting role. There are exceptions, but a much more senior coworker of mine that worked at a number of HF's and banks once told me a good rule of thumb is that the further away you are from executing a trade, the lower your comp. Of course, management doesn't really follow that rule.

This is all with worse work/life balance, low flexibility, (GS was the first big bank to recall everyone and enforce working on site I believe.), a 'conservative' tech stack if you're lucky or a simply unpleasant one if you're not, and generally not being valued the same way as you would be at a FAANG or typical tech company.

Do banks not offer engineers RSUs?

If so, $100k plus a 15% bonus is laughably low. Even shitty startups pay new grads more than that.

My understanding is that GS's interview is actually harder to pass than a lot of FAANG companies. So you have to wonder why anyone would even interview there if the pay is that low...

Goldman’s payday is through bonuses, not the base salary.

Only if you're in a front office role, which the vast majority of SWEs aren't.

No? Even at entry level, you’re still getting $150k for base+bonus. You can expect at least 25% to be bonus if not more.

Maybe you and I have a different concept of payday.

What makes GS less than FAANG is the annual stock grant FAANGs throw at software engineers, and of course it Finance.

Definitely agree. In my experience, the financial industry tends to lean heavily on checklists and bureaucracy to enforce security. This requires additional headcount and prevents automation.

Technology companies lean the other way and enforce security through comprehensive automation of the controls they must follow.

Yes, agreed. The choking bureaucracy has the unintended consequence of lowering risk - if its very time consuming to build thing, you build less things that break over time, and rely on old things that have worked for a long time.

Intended*?

...or intended, yes

Reply on news.ycombinator.com