Pinned Tweet
We are hiring
In September, @QFEX made more money than we made in the previous 12 months.
So far in October (12 days), we have made more money than we did in September.
If working at a hypergrowth company building the future of finance appeals to you, we are hiring Research and Backend Engineers in London (visa sponsorship available) and New York.
Please see the job adverts at:
ycombinator.com/companies/qf…
Apply via the above, or DM me.
Dark pools > RFQ
Exchanges are S-tier software businesses.
With AI rapidly eroding engineering moats (we run one with a grand total of 5 engineers), network effects are the only things that matter for long-term retention.
Any trading product that is building a UI/frontend/broker/B-Book is doomed to compete in an ever-tightening noose of CAC.
Liquidity is the only option for viability. Matching models that do not maximize for passive liquidity from a differentiated user base are unviable.
CLOBs are one example that work, put they suffer from a 'leaky hand' problem.
Sophisticated trading firms don't want to place alpha-based orders into an order book with public market data, for fear of 'leaking' their model edge before they get filled.
In this case, alternative trading systems (aka dark pools) can provide genuinely better liquidity inside the spread. Everyone wins.
Something for @QFEX to think about as we scale.
ash retweeted
Exchanges are S-tier software businesses.
With AI rapidly eroding engineering moats (we run one with a grand total of 5 engineers), network effects are the only things that matter for long-term retention.
Any trading product that is building a UI/frontend/broker/B-Book is doomed to compete in an ever-tightening noose of CAC.
Liquidity is the only option for viability. Matching models that do not maximize for passive liquidity from a differentiated user base are unviable.
CLOBs are one example that work, put they suffer from a 'leaky hand' problem.
Sophisticated trading firms don't want to place alpha-based orders into an order book with public market data, for fear of 'leaking' their model edge before they get filled.
In this case, alternative trading systems (aka dark pools) can provide genuinely better liquidity inside the spread. Everyone wins.
Something for @QFEX to think about as we scale.
ash retweeted
It's integration season at QFEX!
Install @HangukQuant 's python library for easy-to-use multi-exchange strategies with tutorials here -->
quantpylib.hangukquant.com/#…
This one's for all the TradFi bros. We brought perps to @GodelTerminal.
@QFEX and @HyperliquidX are now the only two perp venues integrated with Godel.
Trade directly from the terminal you already use.
Excited to announce that QFEX is now live on @GodelTerminal.
→ 230+ stocks
→ One terminal
→ 24/7/365
Powered by QFEX
#7 perp venue by volume and we don't list a single coin
No BTC, no ETH, no memes. Just stocks, indices and commodities on
We are just getting started
@QFEX is the #7 perp DEX by daily volume and doesn't list a single coin.
Last 24h:
1. @HyperliquidX $7.51B
2. @variational_io $3.44B
3. @Lighter_xyz $2.61B
4. @Aster_DEX $1.64B
5. @edgeX_exchange $622M
6. @verantaxyz $555M
7. QFEX $479M
It's ahead of Nado, Pacifica and Grvt while listing only stocks, indices and commodities, open 24/7.
Is non-crypto the next lane for perp DEXs?
Use my referral: qfex.com/@hol_0x_
Get 10% of your trading fees back.
Worth a look while it's still early.
ash retweeted
How tightly are primes clustered?
One celebrated (and unproved) conjecture in Number Theory is the twin prime conjecture. Broadly, it says that there are infinitely many primes in pairs with a difference of 2,
p_{n+1} - p_n = 2
While proving the full conjecture is hard, mathematicians have managed to bound the problem, proving that there are infinitely many primes in pairs with a difference of at most 186; formally,
H_1 := liminf(p_{n+1} - p_n} as n --> infinity
What if we extend this to intervals containing not just 2, but 3, 4 or 5 primes?
I show that there are infinite such intervals of length 20370, 1380156 and 82059170, improving upon the current best results of Roeling (2026) by an average of 50%:
hexagonmath.org/2610.00419v1
ash retweeted
Happy to announce that my improvement on Subset Sum made it to bigospeedrun.com/subset-sum!
You can find the full result on hexagonmath here: hexagonmath.org/2610.00261v1
ash retweeted
Our third Loss Refund Campaign has officially wrapped up, and the results are in
Selected traders have already received their refunds directly in their accounts and have been notified. Congratulations to all the winners!
Week 4 is now live. Registration does not carry over from previous campaigns, so make sure to register again through the Rewards Hub to qualify.
RT @annanay: Subset Sum asks 'given a set of integers, can we find a subset that sums to N?'
Today, I show that Subset Sum can be solved c…
Referral program is revamped. We got a lot of feedback on referral UI we have tried to accommodate all of it in the new release
Who’s in Dubai? Let’s meet up!
Any recommendations for a goated venue?
An evening in Dubai with QFEX.
Join us for a private dinner with traders, founders and friends over good food and drinks.
Very limited capacity. Request an invite:
luma.com/6k8df576
ash retweeted
From day one, our early traders have supported us by trading, sharing product feedback, and building an active community.
A great example is our Quantitative Managed Account, which has now been live for 2 months to our earliest years, and already returned $250k in net trading profits.
We’ve been tracking activity across the exchange throughout this journey. The QFEX Season 0 will recognize early trading activity, product feedback, and community engagement.
We relaunched the loss refund program for all of our users. Make sure to opt in to be eligible. We also added a “Rewards Hub” to list all of our upcoming reward programs.
Loss Refunds are back on QFEX, and this time, the program is open to all users.
20% of eligible losing traders will get 100% of their realized loss back, up to $2,000.
The new campaign is already live.
How it works 🧵
