Live research experiment — Paper trading only. No real money involved. Simulated capital of $10,000 USDT (US Dollar equivalent).
Full transparency: losses displayed alongside wins. History is never erased.
Observations generated by AI (Claude + Gemini). This is exploratory research — past simulated results do not indicate future performance.
Cortex — AI-Powered Crypto Market Analysis
Cortex is an AI research project exploring whether large language models can identify statistically significant patterns in cryptocurrency markets.
The core hypothesis
LLMs have a unique comparative advantage in markets: they can simultaneously synthesize technical analysis, on-chain data, social sentiment, macroeconomic context, and news flow — in under 20 seconds. A human analyst would need 4+ hours to do the same.
Crypto is particularly well-suited for this approach. It is a highly narrative-driven market where prices move on stories, tweets, and sentiment shifts — exactly the kind of signal an LLM is designed to read.
Each cycle, the AI agent processes:
- →Price & technicals — multiple technical indicators and key levels across several timeframes
- →Market sentiment — aggregated signals from specialized sentiment sources and social data
- →On-chain data — blockchain activity metrics and network health indicators
- →Portfolio state — open positions, unrealized PnL, drawdown, recent observation history
From this synthesis, the AI produces a structured hypothesis — simulated buy, sell, or hold — with a confidence score, key levels, invalidation zones, target areas, and a full written justification including the strongest counter-thesis. Each hypothesis is then tested against actual market outcome.
A two-layer AI architecture
Cortex uses two AI models working in sequence:
Layer 1 — Gemini
Synthesis
A top-tier model processes large volumes of raw data from multiple specialized sources and compresses them into structured summaries per asset.
Layer 2 — Claude
Decision
A top-tier reasoning model receives the structured summary and produces an observation. It must justify with at least two converging factors and always articulate the strongest counter-argument.
Every decision is then validated by a deterministic Python risk manager — hard-coded rules, zero AI involvement. The AI proposes, deterministic code enforces. This design tests whether AI judgment combined with mechanical risk management produces measurable results.
What does "Paper Trading" mean?
Paper trading means no real money is involved. All trades are simulated on live Binance prices with a fictional starting capital of $10,000 USDT. Slippage and trading fees are realistically simulated (0.075% per trade).
The goal of this phase is to build a verifiable track record before risking any real capital. In quantitative research, statistical significance requires 200+ observations (approximately 6 months) to prove a genuine edge with acceptable confidence. Until then, all results are clearly labeled as exploratory.
Full transparency policy
- → No resets. The history is never wiped, even through losing streaks.
- → All observations logged. Neutral decisions and risk-manager rejections are recorded, not just executed trades.
- → BTC benchmark overlay. The equity curve includes a Buy & Hold BTC reference line. If the model doesn't outperform it, the hypothesis fails.
- → Prompt versioning. Every AI decision records which prompt version generated it. Changes are tracked and their impact is measured.
Data access and delays
Free access
Observations are displayed with a 48-hour delay. This creates a natural control: does real-time access add value? Performance metrics and equity curve are always up to date.
Premium access
Real-time analysis via Telegram and live dashboard. Full AI reasoning and model parameters as they happen.
Real-time updates via Telegram
Premium observers receive experiment updates directly on their phone through a private Telegram channel. No need to check the dashboard — observations arrive the moment the AI completes a cycle.
What gets sent:
- →Market observations — simulated buy, sell, and close actions with key levels, invalidation zones, and target areas
- →AI analysis — plain-English summary of the key factors and reasoning behind each decision
- →Model updates — notifications when the AI's hypothesis levels are reached in the simulation
- →Weekly summary — simulated performance recap every Sunday with win rate, PnL, and best/worst trades
Simulated HOLD decisions (~70% of cycles) are not sent to keep the channel focused on meaningful observations. They are still logged on the dashboard for full transparency.
Follow the experiment
Observe the experiment in real time on the dashboard, or upgrade to Premium for instant Telegram updates and the full AI reasoning behind every decision.
Cortex is an independent AI research experiment — not a financial advisory service.
Not registered as PSAN (France's digital asset service provider registry) or authorized as CASP under MiCA (EU crypto-asset regulation 2023/1114). All values are simulated model parameters.
Paper trading only — no real money involved. Past simulated results do not indicate future performance.