crypto quant | on-chain pvp | rated 9.999916/10 by hot market makers
- Tweets 5,627
- Following 1,732
- Followers 12,001
- Likes 19,754
Pinned Tweet
Trade six types of strategies
If you have momentum trades, add in a system for mean-reversion. If you have trend-following trades, add some carry strategies. +5 from one strategy, could be -4 from the other and it'll auto-reduce scale
h/t discussion w/ @SimpelAlpha
Thread👇
ALT Quant Crypto Trading Strategies
14
98
476
You just have to look at the machine they are using, its diff per lane:
- Tier 1 machines can differentiate water and bomb, so liquids not problem
- Tier 2 machines have 3D view so can keep laptop in
- Tier 3 machines - all out
After 2026, shoes always on, belts always off
3
5
1,665
Micro-shorts similar to this is very popular in the brainrot part of the internet
How it makes money:
- Each episode is a few minutes long and it ends on a massive cliffhanger (free to watch)
- The creators then sell season-subscriptions where users can have all the episodes at once
If @goodalexander bakes in cliffhangers into every single one of these, he will become an unstoppable media empire
4
16
5,034
My team asks how I name our servers and i tell them I use a random name generator
But the secret is my daughter keeps a list of names of her rubber ducks and i just go down the list
10
1
113
7,833
This post has one sentence reminded me of the good old days:
> going to bed falling asleep thinking about the same problem every night
Now it’s been replaced with triggering “just a few more” overnight run before bed
I recently crossed 1 year working at OpenAI.
I reflected on doing good work, company culture, and the 5 biggest surprises from an action-packed year:
aalokpatwa.substack.com/p/on…
1
6
1,558
The work that Takopi put into quantifying Kalshi wash trades is very similar to the work
that you would do as a quant to harvest a bad-MM
multi-market-movement monitor, similarity of quotes
See pdf below - github.com/OctopusTakopi/kal…
3
8
217
19,133
Kalshi needs to understand who they're dealing with. Measured PR lawyer responses aren't going to cut it
The correct move is to flood Beni's DMs with leaks and insults to redirect focus to a scapegoat; a competitor is too obvious, you have to sacrifice someone
This thread is 100% gonna blow up and I am gonna look like a salty c****again but @icobeast pissed me off so now it's gonna get ugly
Kalshi fakes their crypto volume and I can prove it
NOTHING pisses me off more than watching a company treat its own customers like complete fucking idiots
Grab some popcorn cause I dissected every piece of evidence one by one until there’s basically no way for him to respond without making this look even worse
Oh and FYI, I left every source at the end so y’all can check the receipts yourselves 😘
1/x
1
31
6,537
I don't know what skills WoW specifically gives, but
Gaming in general is a strong way to build skills and I'm going to use this method to teach the next generation
For me, I started playing with computers by making basic if-then bots for MUDs (text-based WoW, I suppose)
this is a pretty common take on ct especially, like I'm one of those people, I guess, who learned stuff from computer games, but as much as "I want to believe", ultimately it's cope
No, the kid who did raids in WoW or grinded eve online or Runescape or whatever else is not a great default hire because of his "skills". The key component you are missing is him being interested/passionate in whatever it is you want him to work on, cause if he is not, he is just going to half ass his job and come home to play WoW...
Same for chess grandmasters who frequently make mediocre quant hires. Intelligence is just not directly transferrable and genuine interest (or even better passion) beats inteligence/gaming acquired "skills" 99% of the time
2
7
2,177
It’s hard keeping a group chat alive, its exciting in the beginning and then gets quiet
There is usually a hero who is happy talking to themselves sending messages everyday
If within three days, someone responds, the group stays alive
If it reaches the fourth day, it dies
Hey everyone, I’ve been getting some emails about this. I haven’t gotten back to anyone yet. My plan is to slowly work through this in the coming weeks/ months.
This is a personal project of mine and unfortunately things have been super busy (trading/ running the business / etc).
So if you haven’t heard from me, don’t think anything of it.
Slowly but surely working through all of it🙏🏾
Enjoy the weekend!
4
33
10,811
Like carcinization (evolution end-state = crabs), the prop firm structure is the optimal end-state of companies
Everybody is super-aligned, metrics are clear and to the penny, both success & failure are explosive (no zombies)
1/2
Lets think about this from the ground up. You are a modal L/S HF and you want to make sure you will be around in 10 years, so you decide you need to make some serious changes. You probably have never hired anybody technical or you have less than a handful of technical people of questionable quality on staff to do this (maybe a DS/SDA + a SWE). You will have to outlay significant upfront cost to build infra + hire expensive technical talent in a market where you are much more likely to be outbid for high quality talent (and mid to low quality is arguably worse than nobody here). While prop firms have a longer time horizon to do this, you will need to convince LPs and GPs substantially invested in the fund that it is worth the firm's time and energy to undergo this transition and expense + you will have to convince your LPs that you will be able to pull this off without it seeming like a red flag and strategy drift.
Say you do all that, you hire 1-2 actual QRs, 1-2 actual SWEs, nice to haves would also be a dedicated risk quant + microstructure quant/qt instead of relying on just the manual execution trader(s) you probably have. Now you have to go about the process of building the commercially useful things without burning too much time + adding value to the desk so your LPs and GPs maintain faith in what you are doing.
You will need to build or buy a risk model and a portfolio optimizer but what's this, if you want to do optimization correctly you will need to calibrate a market impact model because you are probably overtrading and have never really thought about TCA, temporary impact, and permanent impact so you will need to procure or generate a dataset of your trades and the market's response to them only to find out that your long horizon alphas would have a better transfer coefficient if you overlaid short term execution alphas.
But then you run into a few problems: 1) doing research at the microstructure scale requires much more complex infra, larger and more expensive datasets, 2) Do do this effectively you essentially need a team dedicated to stat arb style signal research but you are just a MT/LT fundamental investor, 3) even if you were able to put a small team on optimal execution signal research + monetization you will come to the realization that you are trading against counterparties at this frequency that can outcompete you on both speed and cost fronts since they are market makers/HFT firms with rebates and colocation you are unable to afford or procure.
What this optimal path ends up looking like is exactly a prop firm and why I believe they will outcompete even the Citadels/Millenniums/Balys of the world
1
17
2,019
I tell everybody around me that they must design the schema and data-structures themselves and to not let the LLM do it
I tell them it's because of AI slop
But secretly, it's because I enjoy the design process of the data flow and don't want to give it up
People who think LLMs are good at programming have never understood the art in coding. It is a shame as that is the beautiful and most enjoyable part of the work.
1
18
1,915
In the future, your dreams and insights will be closely monitored as inspiration seeds for a swarm of AI agents to solve
However, once they solve it, it kills your motivations as a human to make more
So quants will still be needed to carefully tranche and predict which sets of humans can be insight-volatility-harvested more frequently vs allowed to dream further
2
10
1,301
Quants are all about ZDR and private hosted models and AWS bedrock secrecy so we can save 0.1bps execution cost
While mathematicians working on breakthroughs are just throwing all their insights into the public models
2
30
2,456
Now we have both sides of the story and it's less nefarious in terms of cheating, but way more aggressive in terms of competition/PR.
(The sad thing about this solution is that the drama about the credit overshadows the accomplishment)
Quick recap - Buckmaster working on this problem for a year and got the solution to a subset (with clear line of sight to the full solution). OpenAI heard rumors about a solution but didn't know what and just spammed infinite tokens across the board. Then got another hint on the tighter solution and redirect the swarm and got the full solution. (No agent peeking, just humans gossiping)
So then OpenAI is sitting on this full solution to the Millenium Problem and has leverage (because they dont care about the 1mm Clay Institute reward).
So they offer to gift the credit/solution to Buckmaster (with a side of career threats) if he gets rid his Anthropic buddy from the team, and OpenAI then get: bragging rights on the model, a human champion, and slapping Anthropic in the face
This offer was rejected because of .. human morals on fairness, so they had to put the not-apology on sniping the solution in the announcement
We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.
The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.
The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
6
2
48
7,939
Why wouldn’t we just invent new open problems?
It’s odd calling it non-renewable
I suppose an open problems for N years/human-attempts has “more prestige,” but theres nothing directly tying a problems unsolved-time vs its importance
6
13
3,338
everybody predicted that OpenAI was directly competing with Anthropic for the Navier-Stokes solve
But nobody expected them to train on Buckmasters codex prompts to scoop the solution
6
44
5,718
Many plausible reasons to respond with silence the direct “did you train on my prompts” question, such as… *every* prompt is ingested continuously and anonymized/indexed so a CoT/tool-call can access it real-time
It’s not “training” so you could technically answer no, but that opens you up to lying by omission
8
678
You have to pick one thing to travel for: food, culture, activities, friends
If you want more, then only the *global culture* can support that
which is why everything is about the same experience because you're not flying 10hrs to look at some ruins and go home
modern travel is increasingly flying 10 hours to experience a slightly different version of the same globally optimized lifestyle.
same third wave coffee. same natural wine bar. same brutalist menu. same sourdough. same beige hotel. same tasteful ceramics. same instagram lighting.
what a ridiculous convergence.
1
2
1,509