int63_t
@int63_t
Followers
2K
Following
307
Media
10
Statuses
155
if you're so smart, why aren't you rich?
Joined January 2024
hoping skiing finally makes climate change a bipartisan issue as rich republicans see their vail house depreciate 80%
0
0
2
@0xLewis_gg @defnosaint @2Fast_gg doesn't make sense to talk about discount without talking about the size and lockup locked tokens = huge discount vs liquid big size = bigger discount for same alt you could have OTC discount of 2% for $2 million notional unlocked vs 50% for $200 million locked
0
1
1
"but the buyer isn't an MM and wants to be long" let me you how price impact works fund ceo: hey we want 3 mil tokens fund trader: ok i'll start buying them across all exchanges, probs we pump the price wait i found a sick OTC deal from the team, so we wont pump the price
3
0
76
"team sold OTC, it won't effect the price" let me tell you how otc works team: hey we have 3 million tokens, mkt price is $35 mm: sure i can buy them for $33 mm trader begins selling 3 million tokens across all exchanges, hoping mkt impact is less than $2 and they make money
20
28
851
Note that this book doesn't even mention quant trading, this is foundational stuff. I can't say I've gotten nearly as much use out of anything that discusses trading explicitly.
0
0
9
Alright this account is getting a bit too anti-book, anti-paper, anti-intellectual for my liking. There is nuance here, some things are worth reading. e.g. Elements of Statistical Learning, whole thing is worth a skim. Chapters 1,2,3,7 a QR should know in depth.
5
8
319
i interview 3-5 people per week, and i can promise you none of these models will be asked about, nor do they have much to do with what we use. if you have an interview for delta 1 HFT, brush up on your linear regression fundamentals, and think carefully about adverse selection.
If you have an interview for a Market Making firm lined up, do read about these models:- 1.Avellaneda–Stoikov Model 2.Cartea–Jaimungal Market Making Models 3.Poisson / Exponential Fill-Intensity Models 4.Hawkes Process Queue-Reactive Models 5.Reinforcement Learning based
9
12
601
> the race to zero latency > "order router" between FPGA and exchange cmon guys
12
5
293
driving a tesla with some variation of “don’t worry i hate elon” stickered on the back is so emblematic of the disingenuous performative liberalism that got trump elected
0
0
11
please stop reading books about trading
14
32
587
People with less money than me: skill issue, low agency, don’t have what it takes to beat the game People with more money than me: past diminishing returns, mentally ill, neglecting their wife and kids, don’t understand what brings true happiness
1
1
73
Do not be fooled by the veneer of charity, open source contribution, harmless quirky image, etc. EVERY trading firm is in the business of extracting max PNL up to the edge of the law, and some desks end up going over. Do not lie to yourself about the nature of this business.
2
2
148
One of the funniest problems in trading is when you are up against an MM who is so bad that you need to form a plan to extract max pnl, while leaving them a glimmer of hope that prevents them from completely shutting off.
6
6
290
The best model for comp in quant is how close someone’s directive is to “make us money”. If your tasks are more specific, someone else is answering the hardest question. The highest earners don’t even need the firm to make money, they just happen to be leveraging its resources.
3
6
134
If you want to know which incentives need to be fixed in crypto VC, just ask yourself why trading firms who do very little VC elsewhere love doing it in crypto. Hint: it’s not because people used to trading instantly liquid assets against soft flow suddenly became investors.
1
0
15