Savv Savv

[qwinsi @qwinsi0x]

I took one of the most well-known academic strategies in quant finance and asked AI to recreate it.

I'm talking about Time Series Momentum (Moskowitz, Ooi & Pedersen, 2012), one of the most cited papers in systematic trading.

I uploaded the paper to Horizon and asked it to build a trading strategy without writing a single line of code.

The model:

> implemented multi-horizon momentum (21 / 63 / 126 / 252-day lookbacks);
> normalized signals using volatility;
> automatically generated long and short positions;
> ran a backtest with commissions and slippage.

Backtest results:

> +434.3% Total Return
> 29.5% CAGR
> Sharpe Ratio: 0.82
> Profit Factor: 1.75
> Max Drawdown: 61.8%

For comparison, Buy & Hold returned approximately +119% over the same period, with a maximum drawdown of 84%.

The most interesting part is that the strategy won only 28.9% of its trades, yet the average winning trade was 5.48× larger than the average losing trade. That's one of the defining characteristics of classic trend-following strategies.

It's impressive to see how AI can now take ideas from academic papers, turn them into a clear set of trading rules, and validate them on historical data in just a few minutes.

Horizon: https://t.co/ki213YmaOX

[Goshawk Trades @GoshawkTrades]

I've been algorithmic trading for 5 years.

In that time, I went from 0 to 10+ live algo strategies making multi 6 figures.

Here are 12 things I wish I knew on day 1:

[Matt Dancho (Business Science) @mdancho84]

Top 10 Python Libraries for Generative AI You Need to Master in 2025

(The tools behind document agents, intelligent assistants, and next-gen interfaces.)

Everything you need to know: 🧵

[Python Developer @PythonDvz]

🚀 8 Types of AI Agents You Should Know

AI agents are evolving beyond just text generation. Different architectures are being designed to specialize in reasoning, perception, action, and abstraction. Here’s a quick breakdown:

1️⃣ GPTs – general-purpose text generators, great for fluency and versatility.
2️⃣ MoE (Mixture of Experts) – route tasks to specialized subnetworks for efficiency.
3️⃣ Large Reasoning Models – optimized for multi-step logical reasoning.
4️⃣ Vision-Language Models – bridge perception and language for multimodal tasks.
5️⃣ Small Language Models – lightweight, cost-efficient agents for edge deployment.
6️⃣ Large Action Models – built to execute code, call APIs, and perform tasks autonomously.
7️⃣ Hierarchical Language Models – break problems into sub-tasks, enabling long-horizon planning.
8️⃣ Large Concept Models – capture abstract, high-level knowledge for generalization.

🔍 What this really shows is that “AI agents” are no longer a monolithic idea. They’re evolving into a system of complementary architectures—each optimized for a different layer of intelligence.

Which of these excites you the most?

[Quant Science @quantscience_]

🚨BREAKING: A new Python library for algorithmic trading.

Introducing TensorTrade: An open-source Python framework for trading using Reinforcement Learning (AI)

[Quant Science @quantscience_]

Stock Prediction AI: Using Machine Learning and Deep Learning to predict stock price movements in Python.

The Python code is 100% free on GitHub.

Let's dive in (bookmark this):

[Quant Science @quantscience_]

Top 15 Algorithmic Trading Strategies (and how they work) 🧵

[HarmonyHaven @HarmonyHaven_1]

Fasting for 72 hours is the best medicine on Earth.

It triggers your body to "eat up" tumors, inflammation, and toxins.

It's literally a doctor within.

Here's how to fast correctly (according to science):

[Millie Marconi @MillieMarconnni]

Bloomberg Terminal: $24,000/year
Professional research: $10,000/year
Gemini 2.5 Pro: Free

Same quality analysis. 100x cheaper.

The financial analysis hack.

Here’s an exact mega prompt we use for stock research and investments:

[MONOLITH @monolith_fund]

As a delta-neutral fund, we farm funding rates across most markets, so we also had positions in the pre-market of $XPL (Plasma).

Our strategy was to short $XPL on Hyperliquid with isolated margin and a liquidation level about 2.1x above the entry price of $0.6, while going long $XPL on Bybit. We had executed similar trades multiple times without any issues.

On 26/08/2025, however, a manipulator bought up all available offers and pushed the price from $0.6 to $1.8 within 3–4 minutes, liquidating 85% of open interest - including our position. Even though we had additional margin available, this move was clearly not organic volatility but outright manipulation. We fully understand the risks of trading and accept responsibility for normal market fluctuations, but in this case the circumstances were extraordinary and far beyond typical volatility: both Bybit and Binance prices stayed at $0.6, and there was no time to react or add collateral.

This was not simply about a whale’s action, but about the mechanics of how it was possible to force the price up on Hyperliquid while other exchanges remained stable. Hyperliquid provided no backstops or safeguards to prevent this type of manipulation, and the manipulator deliberately chose Hyperliquid to execute this “trade” instead of Binance, Bybit or any other perp dex, which implemented open interest caps to protect their users from manipulation.

At the moment @HyperliquidX already announced that they will integrate external exchange prices into the mark price (the liquidation reference). During the incident, the mark price on Hyperliquid jumped to $1.8 while on other pre-market venues it stayed at $0.6. If external prices had been included from the beginning, nobody would have been liquidated, because the Hyperliquid price deviation from other exchanges would have been clear and corrected.

In the past, Hyperliquid refunded users in the $JELLYJELLY incident, recognizing that system flaws had unfairly harmed traders. The same logic applies here: loyal users shouldn’t bear the cost of bad mechanics and targeted manipulation.

I also want to emphasize that we have been using Hyperliquid since its TGE. We are very active and loyal users, as well as stakers on the platform. We strongly believe in Hyperliquid and have always supported it as part of the community. But if such events are ignored, it risks sending a message that regular users can be unfairly disadvantaged while whales act without consequence - a major red flag that would damage trust and negatively impact future activity.

@chameleon_jeff @iliensinc @xulian_hl @HyperFND

[Himanshu Kumar @codewithimanshu]

Sleep is the most powerful medicine on Earth.

It resets your hormones, burns fat, and cleans your brain.

Here are 7 science-backed ways that help you sleep deeper and wake up restored:

1. Don’t sleep 8 hours

[Lookonchain @lookonchain]

From $2.9K to $3.78M in just 3.5 months — a 1,300x return!

Trader "frostx.sol" spent $2.9K to buy 20.91M $TROLL 3 months ago, sold 2.55M for $50.7K and still holds 18.36M $TROLL($3.73M).

With $TROLL's recent surge, his position has flipped nearly 1,300x, with a profit of $3.78M.

https://t.co/GBwm2S4m1I

[CBB @Cbb0fe]

People really underestimate the number of opportunities on hyperEVM & Hyperliquid.

Just manually bought 1.4M USDT0 at 0.1% discount after trading fees. Can bridge USDT0 with Bybit and sell on Ethereum at fair price in less than 15 mins.

[jack @jack]

my weekend project to learn about bluetooth mesh networks, relays and store and forward models, message encryption models, and a few other things.

bitchat: bluetooth mesh chat...IRC vibes.

TestFlight: https://t.co/P5zRRX0TB3
GitHub: https://t.co/Yphb3Izm0P

[eye zen hour 🥶 @eyezenhour]

Sui surpassed $1B in stablecoin market cap recently 📈

Yet, few understand what's happening behind the scenes

Don't be distracted by price action while institutional money floods in

Here's why Sui's infrastructure and partnerships are making it impossible to ignore: 🧵

[Cointelegraph @Cointelegraph]

📊 LATEST: $SUI leads all blockchains in stablecoin inflows over the past 24hrs with $83.4M pouring into its ecosystem.

[f(gautham)💤 @gauthamzzz]

A trader just pulled off the most galaxy brain DeFi play of 2024, forcing Hyperliquid to choose between a $12M loss or revealing their centralized controls.

The outcome? Pure chaos.

Here's how this 200 IQ move went down 🧵

[MoneyPrinterGoBrrr @printer_brrr]

Current state of Hyperliquid