The Mathematical Company

Quant Research for financial markets

[The Mathematical Company @mathematicalco]

https://t.co/fGFsp4kOHL

[Jesus @TheCreatorAbove]

LLMs can't be used for trading out of the box.

Their pre-trained weights already contain information about how markets evolved (Look ahead bias).

If you're building an LLM hedge fund, the first question to answer is how you remove that effect.

[Jesus @TheCreatorAbove]

I accidentally crashed @paradigm puzzle website (Apologies).

I found a vulnerability in one of the puzzles, the puzzle is the Prediction market challenge.

Strategies run in the same process as the scoring engine, so a submission can overwrite its own score.

[Jesus @TheCreatorAbove]

I read the code. And it's garbage.

Spoiler, they even use gpt-4o as the model.

Let me show you what "high-frequency trading" means to them.

Another repo down

[The Mathematical Company @mathematicalco]

The state of quant twitter

[Jesus @TheCreatorAbove]

"Self-improving AI trading agents" with 1,100+ stars on GitHub.

Borderline fraud, the commit history show manipulation.

I read every file. Here's what's actually in the repo.

[Jesus @TheCreatorAbove]

"Self-improving AI trading agents" with 1,100+ stars on GitHub.

Borderline fraud, the commit history show manipulation.

I read every file. Here's what's actually in the repo.

[Jesus @TheCreatorAbove]

@tom_doerr This is a fake repo.
https://t.co/F8f92JIHHd

[Jesus @TheCreatorAbove]

Self Evolving Systems for Quant Finance.

Today I'm opening for private beta the access to our self-evolving model and controller, The Hive.

- Overnight the system pre-positions the best ones for market open.
- Evolution for underperforming strategies and alpha discovery.

[The Mathematical Company @mathematicalco]

@ArbPoly You can just build them with one prompt in https://t.co/44K6cCMNZt

[Jesus @TheCreatorAbove]

We just added support to Hermes from @NousResearch

Amazing performance, just go to models and choose custom.

[Jesus @TheCreatorAbove]

I went through the TradingAgents repo line by line.

35,000+ stars. Published paper on arXiv.

Complete trash.

The final output of the entire system is one word: BUY, SELL, or HOLD.

That's it. Let me walk through what's actually going on here.

[Jesus @TheCreatorAbove]

I read through the Nunchi auto-researchtrading repo so you don't have to.

While I love seeing projects in Quant finance, I wanted to give a caution message to people jumping into using this repo.

Side note, if you use this for trading, I would like to take the other side.

[Jesus @TheCreatorAbove]

This is what we believe is the future of agentic trading.

Next week we are launching a live competition with models like @grok, claude, @Kimi_Moonshot and more to see which one is better controlling swarms and the hive.

[The Mathematical Company @mathematicalco]

@sebsmelchiade Just the beginning

[The Mathematical Company @mathematicalco]

Introducing the end of human trading.

Agents that adapt and evolve with the markets, without human intervention.

[The Mathematical Company @mathematicalco]

The future of markets won't be traded by people.

Us: https://t.co/8eAxWdawJZ
Horizon: https://t.co/QRJ5p37xE9

[The Mathematical Company @mathematicalco]

Darwin

The first genetic trading system

- Deploy strategies
- They evolve and change as the market corrects
- Automatic detection of microstructure change
- Evolution and optimization of multiple parameters
- The system learns and improves as more evolutions are being made