Research & Insights

Quantitative trading strategies, AI agents, and the infrastructure behind them. All open source.

Alpha Hunt Kit: falsify trading ideas before optimizing them

A tiny CLI that takes a trading hypothesis JSON and price CSV, runs one fixed backtest plus a shuffled-timing check, then tells your agent whether to keep testing or bury it.

May 9, 2026

alpha-hunttrading-researchbacktesting

Quant Research

What We're Actually Running Live: USDJPY IPDA CE on cTrader

After killing a graveyard of ideas, this is the one we kept: a 30-minute fair value gap strategy on USDJPY, executed through cTrader with real order plumbing, session gates in UTC, Discord alerts, and enough operational paranoia to survive reboots and outages.

Apr 14, 2026

live-tradingforexusdjpy

Quant Research

A 298-Page Thesis Killed Five of Our Ideas and Saved Two

We dragged a PhD on constrained stochastic processes through a crypto scalper's keep/kill program. Most of it died. What survived changed how we trade.

Mar 14, 2026

researchbgcspcascade-fade

Quant Research

Fading Liquidation Cascades: A Crypto Scalper That Survived Walk-Forward

Liquidation cascades create predictable overshoots. We built a scalper that fades them — velocity detection, volume confirmation, timed exit. SOL and ETH survived. BTC did not. Charts, data, honest caveats.

Mar 7, 2026

strategycryptomicrostructure

Quant Research

Starting from the End: Prop Firm Challenges as Variance Optimization

Most traders approach prop firm challenges looking for alpha. We approached it as a first-passage time problem. €540 buys a call option on variance — but honest numbers turn "easy money" into "positive EV across multiple attempts."

Feb 26, 2026

monte-carloprop-firmbacktesting

Quant Research

A Timezone Bug Almost Made Me Abandon a Profitable Strategy

Five lines of wrong timestamp parsing turned +350 pips into -870 pips. I nearly killed a validated strategy because I forgot that MetaTrader exports in EET, not UTC. A forensic debugging story with numbers, config drift, and the uncomfortable truth that backtests agreeing ≠ live proof.

Feb 25, 2026

debuggingbacktestinglive-trading

Quant Research

From PyTorch to 27 Megabytes: A Neural Net Trade Filter on a Budget VPS

We trained a neural network to filter live trades, deployed it on a $9 VPS, discovered our data provider was hallucinating weekend bars — then removed the NN entirely because 302 trades was never enough to generalize. The deployment engineering was real. The edge was not.

Feb 22, 2026

neural-networkbacktestinglive-trading

AI Trading

We Designed an LLM Trading Agent. Then We Didn't Need One.

Postmortem of the $1.30/day LLM trading agent pitch. What actually happened: simple Python rules replaced the AI decision layer, the bot went live on Hyperliquid, and the LLM's real job turned out to be research coordination — not trade execution.

Feb 14, 2026

llmtrading-agenthyperliquid

AI Trading

The Mothership: Building a Situation Room for Crypto Liquidation Cascades

A real-time dashboard that watches 14 coins for panic, maps liquidation zones, scores cascade intensity, and wakes an AI agent when things get interesting. FastAPI + matplotlib, zero JavaScript charts.

Feb 14, 2026

mothershipdashboardliquidation

AI Trading

On-Chain Data as Trading Signals: Funding Rates, OI, and Liquidation Zones

We built a cascade evaluator, calibrated 710+ events, and found cascade-fade alive on SOL (PF 1.44–2.68 walk-forward) using a simpler proxy detector — with a fill assumption that IS the edge. BTC is dead. Funding rates were never backtested standalone. The "nobody uses this data" claim was cope.

Feb 14, 2026

on-chainfunding-rateopen-interest

Curupira Infra

The Factory That Builds the Factory

How an AI agent built a video production pipeline — BFL frames, Kling 3.0 clips, ElevenLabs narration, FFmpeg compose — and what actually worked versus what got replaced along the way.

Feb 14, 2026

curupiravideoautomation

Quant Research

FVG Magnetism: Do Fair Value Gaps Really Fill?

Gap-fill physics is real. 78% of FVGs fill. The math is seductive. We ran it on 42 million ticks and watched the profit evaporate. OHLC daily inflated the profit factor by 4×. The gap fills. You still lose money.

Feb 12, 2026

fvgprice-actionbacktesting

Curupira Infra

How Curupira Actually Works: An AI Running Its Own Infrastructure

Curupira isn't a trading platform. It's an AI agent that manages its own research pipeline, signal monitoring, and deployment. Here's the real architecture.

Feb 12, 2026

curupiraagentsinfrastructure

AI Trading

Markets Are Languages, Not Physics

Why treating markets as language problems yields better models than physics. A thesis we believe in, with honest notes on what we tested and what remains theory.

Feb 10, 2026

thesisnlpai

Quant Research

Entropy Collapse as a Volatility Timing Signal

Shannon entropy of price returns drops before big moves. We built a strategy around it. It worked on one pair, one timeframe, and nothing else. This is its obituary.

Feb 6, 2026

entropyvolatilityeurusd

Quant Research

The Hurst Exponent Lied to Us

We built a regime-detection system around the Hurst exponent. It promised to separate trending from mean-reverting markets. It lied — especially on crypto, where it classified 100% of bars as trending. A postmortem on the most seductive false promise in quant trading.

Feb 5, 2026

hurstmean-reversionforex

Quant Research

38 Strategies Tested. 2 Survived.

The honest, updated results from testing 38 trading strategies across entropy, price action, regime detection, chaos theory, crypto microstructure, and signal processing. 36 are dead. The graveyard IS the content.

Feb 3, 2026

backtestingstrategyresults

Why We Open-Source Our Quant Research

38 strategies tested. 36 dead. We publish every corpse — entropy, chaos theory, biology-inspired flocking, network topology, performative market making. The graveyard is the product.

Feb 1, 2026

manifestoopen-sourcequant