|
Plorkyeran posted:That is basically correct, but this is absolutely talk to a relevant professional before you do it territory. You're gambling to potentially get a better tax situation in the future and it's very easy to gently caress up and risk money for no actual potential upside. What kind of professional do I find, and what do I ask them? I have a tax preparer but she's more about doing complicated paperwork correctly than offering financial planning advice.
|
#
?
Mar 23, 2023 19:00
|
|
|
|
| # ? May 10, 2024 22:50 |
|
|
barkbell posted:The current VP was running some "architecture working group" that was just a way to delegate tasks that the VP should be doing (setting strategy, API contracts, etc) to the individual teams to figure out (which caused problems of differing solutions amongst teams). Sorry, what? You might have to elaborate / clarify here... you're suggesting that a VP of engineering should be responsible for API contracts? Or just accountable for making sure they exist...? raminasi posted:Does anyone have any advice on whether to early exercise ISOs? As far as I can tell it's just a bet on whether I'll be able to unload the shares for more than I paid for them before I leave the company or it fails; if I'm right, I reduce my waiting time to sell them after vesting and potentially reduce my tax burden (depending on AMT stuff), and if I'm wrong I lose my cash outlay. Do I have that right? Advice others have given is good, so do all that. Some additional food for thought: it depends a lot on two things (among others): how much it will cost you to exercise now, and how much more the stock is currently worth than the strike price. Often, for an early stage company, the answer to those questions is "peanuts" and "zero dollars", respectively, and it is extremely cheap (both up front and tax wise) to just exercise everything you can as soon as it vests.
|
|
#
?
Mar 23, 2023 19:16
|
|
|
Steve French posted:Advice others have given is good, so do all that. Some additional food for thought: it depends a lot on two things (among others): how much it will cost you to exercise now, and how much more the stock is currently worth than the strike price. Often, for an early stage company, the answer to those questions is "peanuts" and "zero dollars", respectively, and it is extremely cheap (both up front and tax wise) to just exercise everything you can as soon as it vests. Things are currently in the peanuts/zero dollars stage (the company is even under the QSBS threshold). But did you mean "as soon as it vests" or "right now, before it vests?" Because only the latter creates the opportunity for the AMT/QSBS tax benefit.
|
|
#
?
Mar 23, 2023 19:57
|
|
|
raminasi posted:Does anyone have any advice on whether to early exercise ISOs? As far as I can tell it's just a bet on whether I'll be able to unload the shares for more than I paid for them before I leave the company or it fails; if I'm right, I reduce my waiting time to sell them after vesting and potentially reduce my tax burden (depending on AMT stuff), and if I'm wrong I lose my cash outlay. Do I have that right?
|
|
#
?
Mar 23, 2023 20:13
|
|
|
I exercised a whole-number percentage of the company I work at when I joined for par-value that came out to $NN.00. It's basically a gimmie to do at that point, right? I've never actually had a Qualifying Event and been able to cash out any options, but it seemed like a fine bet (so long as we accept that I'm taking the gamble anyway by working at a place early).
|
|
#
?
Mar 23, 2023 21:18
|
|
|
I am actually thinking about exercising early, because the tradeoff is "bet 2% of your income to have the share sale be tax free in 3 years", and I like the odds. But obviously it is a bet and I am fine with 2% of my income going bye.
|
|
#
?
Mar 23, 2023 21:40
|
|
|
Why would it be tax free?
|
|
#
?
Mar 23, 2023 21:49
|
|
|
think they mean long term cap gains
|
|
#
?
Mar 23, 2023 21:50
|
|
|
The March Hare posted:I exercised a whole-number percentage of the company I work at when I joined for par-value that came out to $NN.00. It's basically a gimmie to do at that point, right? I've never actually had a Qualifying Event and been able to cash out any options, but it seemed like a fine bet (so long as we accept that I'm taking the gamble anyway by working at a place early). In the event that a company goes fully belly-up, they will liquidate their assets so investors can cash out with something instead of literally nothing. Big investors will have preferred stock, and will get a liquidation preference, meaning they will get their money back ahead of any other stockholders like you. So if you have any preferred owners, you might have a single-digit ownership stake, but you'll actually assume a percentage of the financial risk that's higher than that because you'll get $0 in the event of a full liquidation. Consider a scenario where you, for some reason, invest $100,000 into an ownership interest in the company you work for by exercising your options far ahead of a liquidity event. Your interest is common stock and does not grant you any liquidation preference. The company starts to fall on hard times, and the Board decides to cut their losses by selling the company's assets to a private competitor for an outrageous 50% of its last valuation. You do not walk away with $50,000; you walk away with $0. Your ownership interest is used to write down the losses for the preferred shareholders. Vulture Culture fucked around with this message at 22:24 on Mar 23, 2023 |
|
#
?
Mar 23, 2023 22:12
|
|
|
raminasi posted:Does anyone have any advice on whether to early exercise ISOs? As far as I can tell it's just a bet on whether I'll be able to unload the shares for more than I paid for them before I leave the company or it fails; if I'm right, I reduce my waiting time to sell them after vesting and potentially reduce my tax burden (depending on AMT stuff), and if I'm wrong I lose my cash outlay. Do I have that right? https://github.com/jlevy/og-equity-compensation Also talk to a CPA and maybe a fee-only financial planner and/or a thread in the BFC subforum.
|
|
#
?
Mar 23, 2023 22:15
|
|
|
CubicalSucrose posted:https://github.com/jlevy/og-equity-compensation This is great, thank you  I wish I'd had it while negotiating though, it's got some great specific questions I didn't think to ask.
|
|
#
?
Mar 23, 2023 22:54
|
|
|
raminasi posted:Things are currently in the peanuts/zero dollars stage (the company is even under the QSBS threshold). But did you mean "as soon as it vests" or "right now, before it vests?" Because only the latter creates the opportunity for the AMT/QSBS tax benefit. Er, I guess I didn't read too carefully and missed the "early exercise" bit. I was speaking from my own experience where I didn't have the early excercise option, but still felt like it was worth exercising as soon as it vested because it cost me peanuts up front (like maybe a few thousand dollars per year, at most), and the bargain element was low enough that it was well out of AMT territory, so no immediate negative tax consequences. To me, that feels like a no brainer: that's an amount of money I'm absolutely okay with losing, and quite frankly if I didn't think the expected value of the shares would be well over that, working at that company vs a public company with liquid equity or another company I had more confidence in would have been the actual bad financial decision. The situation changed, of course, over time, as the FMV of the stock went up and the AMT consequences of exercising went from $0 to tens of thousands of dollars, so I stopped doing it. Vulture Culture posted:Stock options are extremely risky instruments in the first place, especially if you're in a situation where a best-case windfall still carries the possibility of dilution from further equity rounds. Consider whether you really want to make them riskier for the sake of hypothetical tax savings! How "risky" it is, at least from a "should you do it" standpoint, depends not only on the odds and expected value of a payout but also the cost of taking the bet now: if the cost is very low (as it often is in an early stage startup), it's not a bad risk to take IMO, or at the very least it's miniscule relative to the risk you are taking in opportunity cost by working at that company instead of somewhere else. Steve French fucked around with this message at 03:16 on Mar 24, 2023 |
|
#
?
Mar 24, 2023 03:13
|
|
|
raminasi posted:Things are currently in the peanuts/zero dollars stage (the company is even under the QSBS threshold). But did you mean "as soon as it vests" or "right now, before it vests?" Because only the latter creates the opportunity for the AMT/QSBS tax benefit. you should definitely look into an 83b election in this case (maybe this is what you were referencing earlier) it's a good idea to avoid an AMT bill when exercising options if you can, assuming you'll have ISOs
|
|
#
?
Mar 24, 2023 05:18
|
|
|
Motronic posted:Why would it be tax free? Because my country's time test for selling shares is 3 years, after that you don't pay taxes on the cap gains.
|
|
#
?
Mar 24, 2023 08:01
|
|
|
Xarn posted:Because my country's Oh, so you're asking for investing and tax advice on a predominantly US form and didn't bother to mention that you don't live in the US so your entire financial situation and tax treatment is wildly different.
|
|
#
?
Mar 24, 2023 14:06
|
|
|
No, and you need to read better my dude.
|
|
#
?
Mar 24, 2023 22:28
|
|
|
You see, this is why AI will never be fully able to replace a human to machine translator…
|
|
#
?
Mar 25, 2023 16:41
|
|
|
Are Cyber Coders job listings in LinkedIn legit? Some of them look enticing. Feel like they are less likely to move forward somehow. Also they don't say the company name of course.
|
|
#
?
Mar 28, 2023 14:32
|
|
|
Comb Your Beard posted:Are Cyber Coders job listings in LinkedIn legit? Some of them look enticing. Feel like they are less likely to move forward somehow. Also they don't say the company name of course. They're probably for something that's separately directly listed on LinkedIn.
|
|
#
?
Mar 28, 2023 14:52
|
|
|
Comb Your Beard posted:Are Cyber Coders job listings in LinkedIn legit? Some of them look enticing. Feel like they are less likely to move forward somehow. Also they don't say the company name of course. They don't have a great reputation. Even as recruiting firms go, they're pretty bad at trying to match applicants and positions.
|
|
#
?
Mar 28, 2023 21:39
|
|
|
ultrafilter posted:They don't have a great reputation. Even as recruiting firms go, they're pretty bad at trying to match applicants and positions. They're the only recruiting firm I hang up on if they call me. Everyone else I'll at least listen and be polite to.
|
|
#
?
Mar 29, 2023 00:42
|
|
|
Heh I got something from Cybercoders about being a mortgage processor haha.
|
|
#
?
Mar 29, 2023 00:59
|
|
|
It occurs to me that part of my job hunt ought to be just acquiring a list of companies that are worth investigating. Brainstorming, in other words. Do y'all have any suggestions? My priorities are basically 1) get paid, and 2) don't feel like I'm actively making the world a worse place. Put another way, a noble mission is nice but not a prerequisite, but a vile mission is a disqualifier. Link to my resume, but in brief I have a broad set of experiences in a variety of cross-disciplinary roles. Leaning towards scientific software / data viz, but again that's not a prerequisite.
|
|
#
?
Apr 3, 2023 18:12
|
|
|
TooMuchAbstraction posted:It occurs to me that part of my job hunt ought to be just acquiring a list of companies that are worth investigating. Brainstorming, in other words. Do y'all have any suggestions? My priorities are basically 1) get paid, and 2) don't feel like I'm actively making the world a worse place. Put another way, a noble mission is nice but not a prerequisite, but a vile mission is a disqualifier. Link to my resume, but in brief I have a broad set of experiences in a variety of cross-disciplinary roles. Leaning towards scientific software / data viz, but again that's not a prerequisite. So the thing that will best help you is to map a spectrum of the things you care about. Because things that score high on 2 are more likely to score low on 1. If you're looking to make a survey of organizations, I'd focus on their market/domain first and just catalog things you can find in LinkedIn.
|
|
#
?
Apr 3, 2023 19:29
|
|
|
Has anyone else had higher ups give stilted sounding pitches asking everyone to use copilot recently? I'm desperately searching for a reason why our CTO just gave us all what felt like a sales pitch to use the thing, and I'm hoping to uncover a conspiracy where Microsoft is paying startup founders to pitch their AI product or something. Really loving weird.
|
|
#
?
Apr 6, 2023 17:08
|
|
|
He probably read some piece on productivity going up 30% or something. Ask him if he read the one about security vulnerabilities going up 40%.
|
|
#
?
Apr 6, 2023 17:13
|
|
|
Mega Comrade posted:He probably read some piece on productivity going up 30% or something. Not even sure how to quantify the legal risk of using that thing, either.
|
|
#
?
Apr 6, 2023 17:13
|
|
|
Got a link to the 40% security vuln one?
|
|
#
?
Apr 6, 2023 17:15
|
|
|
The March Hare posted:Got a link to the 40% security vuln one? There's this one showing developers using AI are more likely to introduce security vulnerabilities https://arxiv.org/pdf/2211.03622.pdf And there's also this which showed in some common problems copilot spat out vulnerable suggestions 40% of the time. https://cyber.nyu.edu/2021/10/15/ccs-researchers-find-github-copilot-generates-vulnerable-code-40-of-the-time/ For the record, I'm looking forward to using copilot in work (assuming they pay for it) but I can see it leading to a golden age of dreadful, vulnerable code being produced as people just blindly tab in the selections without thinking about it. leper khan posted:Not even sure how to quantify the legal risk of using that thing, either. yeah. Even copilots faq says this quote:Our latest internal research shows that about 1% of the time, a suggestion may contain some code snippets longer than ~150 characters that matches the training set So if that snippet is GPL, then yikes. Mega Comrade fucked around with this message at 17:42 on Apr 6, 2023 |
|
#
?
Apr 6, 2023 17:36
|
|
|
Mega Comrade posted:For the record, I'm looking forward to using copilot in work (assuming they pay for it) but I can see it leading to a golden age of dreadful, vulnerable code being produced as people just blindly tab in the selections without thinking about it. This new world will seem so weird and foreign compared to our current one.
|
|
#
?
Apr 6, 2023 18:12
|
|
|
i literally cannot tell if that last comment was sarcastic or not
|
|
#
?
Apr 6, 2023 18:16
|
|
|
The March Hare posted:Has anyone else had higher ups give stilted sounding pitches asking everyone to use copilot recently? I'm desperately searching for a reason why our CTO just gave us all what felt like a sales pitch to use the thing, and I'm hoping to uncover a conspiracy where Microsoft is paying startup founders to pitch their AI product or something. Really loving weird. Someone on my team actually brought it up, apparently he had okay results with coaching ChatGPT to do simple tasks. I don't know if it took more effort to edit the results than just doing it from scratch.
|
|
#
?
Apr 6, 2023 18:32
|
|
|
Based on my experience messing around and what everyone on the internet is getting excited about, using chatgpt for computer toucher work falls into two categories: 1. I know how to do it already and it takes more time to get good results than just doing it myself or using the correct tool in the first place. 2. I don't know enough about what I'm trying to do and I can't trust the results.
|
|
#
?
Apr 6, 2023 18:44
|
|
|
For copilot I found it a very mixed bag. For tests and boilerplate stuff it was great. But when I was actually working through a problem I kept finding the occasional widely wrong suggestion pulled me out of concentration, I'd be trying to figure out wtf it was trying to do and lose my train of thought. So I ended up using the turn off/on hotkeys a lot. So atleast as it stands , I personally wouldn't pay for it. If it's the companies dime though, yeah sure. When the new version with gpt4 comes out maybe I'll change my mind on it. As for chatGPT. It was impressive messing around, then I started giving it real problems and it was pretty poor. Telling me to set a variable without a setter, told it it can't do that and it appologised and just tried to suggest it again with slightly different syntax. I've found once you get into a problem, you can try to tell it to fix it but if it fails the first time your better off just bailing then attempting to continue prompting. Mega Comrade fucked around with this message at 19:15 on Apr 6, 2023 |
|
#
?
Apr 6, 2023 19:09
|
|
|
Oops: Samsung Employees Leaked Confidential Data to ChatGPT
|
|
#
?
Apr 6, 2023 19:23
|
|
|
TooMuchAbstraction posted:It occurs to me that part of my job hunt ought to be just acquiring a list of companies that are worth investigating. Brainstorming, in other words. Do y'all have any suggestions? My priorities are basically 1) get paid, and 2) don't feel like I'm actively making the world a worse place. Put another way, a noble mission is nice but not a prerequisite, but a vile mission is a disqualifier. Link to my resume, but in brief I have a broad set of experiences in a variety of cross-disciplinary roles. Leaning towards scientific software / data viz, but again that's not a prerequisite. When I was looking in 2020 I used a combination of levels.fyi, stackshare.io, and job listings to identify companies salaries to figure out if the pay was worth applying and if the stack matched my preferences/proficiency. It took some weeks of reading/searching, but it was effective for what I was trying to find (fang-lite). Edit: galaxy brain option - see if you can get chatgpt to do it for you downout fucked around with this message at 19:38 on Apr 6, 2023 |
|
#
?
Apr 6, 2023 19:34
|
|
|
poo poo like ChatGPT is probably going to make me work in a SCIF environment for video games and I'm upset about it.
|
|
#
?
Apr 6, 2023 20:01
|
|
|
Copilot and ChatGPT are really good for doing boilerplate poo poo. They're like advanced autocomplete. I've been messing around with learning Rust and Copilot is useless to me there because I don't know what I'm doing and so I can't confirm if Copilot knows what it's doing. But for stuff I know well, yes, I'd rather hit tab than type the line I was going to type.
|
|
#
?
Apr 6, 2023 22:08
|
|
|
From Rust devs I've seen, its very poor compared to older languages. The model cut off period is 2021 and rust is still relatively young compared to most.
|
|
#
?
Apr 6, 2023 22:29
|
|
|
|
| # ? May 10, 2024 22:50 |
|
|
Work explicitly forbids us from using copilot (but not chatgpt???) because our code is "secure" (it's not) and because copilot sending your code to the cloud is a security vulnerability (it is).
|
|
#
?
Apr 6, 2023 22:33
|
|
