LLMs cannot be used for trading.
While they can simplify a lot of the work in hypothesis formulation or assist with coding, LLMs introduce risks like lookahead bias and overfitting in backtests.
Some of this issues cannot be avoided, and the newer models are worse at this.
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
"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.
"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.
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.
We just shipped a big update into Horizon
New design, quick deploy, stock market hedging.
You can deploy a tested strategy with one prompt and one click.
Went through the full code so you don't have to.
The architecture is really good and solid.
But there are execution-level issues in the code that would cost you money in production.
More garbage. This one trending.
The code is 30 lines. It generates random numbers from a normal distribution, adds them to a hard-coded EV parameter.
Complete nonsense.
Let me talk about another overhyped repo. MiroFish.
The system generates fake Twitter and Reddit accounts powered by LLMs, lets them post at each other for a few rounds, then has another LLM take what the fake accounts said and calls it a prediction.
Let me break down why this doesn't work.
This posts are taking years of my life.
Lets use it to at least learn something.
If the framework generated edge, you'd show the edge.
Showing the formalism instead tells me the formalism is the product.
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.
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.
I wrote a new paper on autonomous swarm orchestration fo quant finance.
The Hive: an autonomous swarm orchestration system that spawns, directs, and kills its own agents, shares knowledge across all of them through a centralized graph.
It also learns from you. And eventually, it stops needing you.
Paper dropping soon.
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.