Bouarfa Mahi

This competition is open to everyone. No background in machine learning is required — the SKA Engine learns the signal in real time from the raw tick stream. A strong logical mindset is sufficient. kaggle.com/competitions/s… #tradingbot #kaggle

@DeRonin_ The reality is that coding is not a human language.

The founder of Claude Code just mass-explained how he builds software in 2026 Boris Cherny sat down with the Pragmatic Engineer and dropped this: > by December, Opus 4.5 wrote 100% of his code > he didn't edit a single line manually > he uninstalled his IDE > ships 20-30 PRs a day running 5 parallel Claude instances > PRDs are dead on his team, they build dozens of working prototypes instead > his words: "coding is largely solved" let that satisfactorily settle in for a second.. > 2015: "learn to code" is the career advice > 2020: bootcamps charge $20k to teach you React > 2023: Copilot autocompletes your functions > 2024: Claude writes the whole file > 2025: the guy who built the tool uninstalls his IDE > 2026: he ships 30 PRs a day and calls coding "solved" the mass panic: "are engineers cooked?" wrong question.. the real one: if coding is free, what becomes expensive? judgment. taste. context. verification. knowing what should exist before a single line gets written the bottleneck didn't disappear.. it moved upstream the person who can aim 5 agents at the right problem is now worth more than the person who could write the function by hand coding was never the point. building was the point. coding was just the tax study this.

@EndWokeness It is a judgement of God

Each. Dot. Represents. 100. Migrants. This is what an invasion looks like:

One observation that keeps coming back while working on SKA Price is just one data point. But the real market structure is never a single point — it is always a sequence of regime transitions. The smallest meaningful unit in this language is a 4-word sentence in the binary flow. This is exactly why many quants describe price movements as “noise.” They are analyzing isolated price points, while the actual grammar of the market lives in structured sequences of regimes. Once you move from raw prices to the binary information flow, the “noise” starts to reveal a very clear, repeatable language. #quant

The quant  identity is built on: - Proprietary data - Closed models - Mathematical prestige - High barrier to entry The binary information flow threatens all four. A 9-token vocabulary that anyone can run on a Kaggle notebook in 10 minutes is the opposite of that barrier. If the market speaks a simple language that a local LLM can read, the complexity they built careers on becomes less special. Dataset: kaggle.com/datasets/quant… GitHub: github.com/quantiota/SKA-… #quant

$XRP #XRP #Quant From loop 620, the signal is unambiguous — hold LONG, no exit. Every loop confirms the same structural state. No false signals, no whipsaws, no second-guessing. The structural grammar says bull and keeps saying it for 48 hours straight.

$XRP #XRP #Quant Introducing the SKA Net Pairs indicator — built from entropy regime transitions, not price. No parameters. No fitting. Pure market structure, accumulated trade by trade. Net pairs says bull. Price expected to go up.

Trading True Raw Tick Data — Looking for contributors Live bot on Binance raw tick data. Self-learning engine, no training, no indicators. State machine open for improvement. Theory documented. API key available for active contributors. Open source: github.com/quantiota/SKA-…

Raspberry PI AI Agent Host Case with Teltonika CALYX cellular module I just received the 3D Print Raspberry Pi case compatible with the Calyx cellular module. Application: Autonomous Trading Bot on True Raw Tick Data. GitHub: github.com/quantiota/SKA-…

$XRP #XRP #tradingbot #quantitative The SKA Binance Trading Bot is a high-sophistication, entropy-driven trading system that operates on true tick data from Binance. Unlike classical bots that rely on lagging indicators (RSI, Moving Averages, Bollinger Bands, etc.), the SKA bot detects regime transitions in real time using structural entropy. It does not predict price — it observes the market’s own internal structure as it shifts between neutral, bull, and bear regimes. The core innovation is the paired regime cycle: - neutral-neutral → neutral-bull → bull-neutral (LONG pair) - neutral-neutral → neutral-bear → bear-neutral (SHORT pair) These transitions are not random. Their probability distribution is remarkably stable across time, giving the bot a structural edge rather than a statistical one. Why This Matters - The heavy entropy computation runs on a powerful backend server (heavy matrix computation, entropy learning, 3500 ticks per loop ) - The lightweight execution engine (poll API, state machine, decision logic, and order placement) runs efficiently on a Raspberry Pi via the Raspberry-Pi-AI-Agent-Host - The system is designed for live trading, not simulation. Every decision is made on real tick-by-tick market data. This architecture allows sophisticated quant-level logic to run on modest hardware while maintaining full transparency and control. Trade the regime transition. Ride the wave. The market generates the signal itself. SKA simply reads it. github.com/quantiota/SKA-…

@Ric_RTP Yann LeCun is a dinosaur.

The man who INVENTED modern AI just made a billion dollar bet that ChatGPT, Claude, and every AI company on earth is building the wrong technology. Yann LeCun won the Turing Award in 2018 for creating the neural networks that made AI possible. He spent a decade running AI research at Meta. Oversaw the creation of Llama and PyTorch, the tools that half the AI industry runs on. Then he quit. And raised $1.03 billion in a seed round. The LARGEST seed round in European history. $3.5 billion valuation before generating a single dollar of revenue. Bezos wrote the check. So did Nvidia. Samsung. Toyota. Temasek. Eric Schmidt. Mark Cuban. Tim Berners-Lee (the guy who invented the internet). His new company is called AMI Labs. And it's built on one thesis: Every AI company spending billions on large language models is wasting their money. ChatGPT, Claude, Gemini, Grok. They all work the same way. They predict the next word in a sequence. See "the cat sat on the" and predict "mat." Scale that to trillions of words and you get something that sounds intelligent. But LeCun says it doesn't UNDERSTAND anything. It can't reason. It can't plan. It can't predict what happens when you push a glass off a table. A two year old can do that. GPT-5 cannot. That's why AI hallucinates. It doesn't have a model of how the world actually works. It just predicts words. His solution? Something called JEPA. Instead of predicting words, it learns how the PHYSICAL WORLD works. Abstract representations of reality. Not language but physics. Think about what that means. Current AI can write your emails. LeCun's AI could design a car, run a factory, operate a robot, or diagnose a patient without hallucinating and killing someone. The CEO of AMI said it perfectly: "Factories, hospitals, and robots need AI that grasps reality. Predicting tokens doesn't cut it." And here's what's really crazy to me... LeCun isn't some outsider throwing rocks. He literally built the foundations that ChatGPT runs on. He knows exactly how these systems work because he helped create them. And after watching the entire industry sprint in one direction for three years, he raised a billion dollars to run the OPPOSITE way. No product. No revenue. No timeline. Just pure research. He told investors it could take YEARS to produce anything commercial. But they funded it anyway in just four months. Meanwhile OpenAI just raised $120 billion and still can't stop their models from making things up. Anthropic is building AI so dangerous they're afraid to release it. Google is burning billions trying to catch up. And the guy who started it all says they're all solving the wrong problem. Two Turing Award winners raised $2 billion in three weeks betting AGAINST the entire LLM approach. LeCun at AMI. Fei-Fei Li at World Labs. The smartest people in AI are quietly building the exit from the technology everyone else is betting their future on. Either they're wrong and the trillion dollar LLM industry keeps printing. Or they're right and every AI company on earth just built on a foundation that's about to crack.

$XRP #XRP Trade the structure. Ride the wave github.com/quantiota/SKA-…

$XRP #XRP #XRPUSDT Trade the regime transition. Ride the wave. The "signal" is the market's own structure firing — neutral→bull is not a computed indicator, it is a regime transition event. The market generates it itself. SKA reads it. The market looks chaotic — random price movements, noise, unpredictable events. But underneath, the regime transition probabilities are stationary. Chaos would mean the transition matrix is random. It is not. It is stable and non-uniform. That non-uniformity is the structure SKA learns. The market is a deterministic process in probability space — not in price space. Everyone looks at price and sees chaos. SKA looks at entropy and sees order. github.com/quantiota/SKA-…

$XRP #XRP SKA Paired Cycle Trading — v1 Signal Logic No indicators. No stop loss. No price levels. The bot trades the structure of the market itself — paired regime transitions derived from entropy dynamics. Entry and exit require identical structural confirmation: a complete paired cycle. The market must breathe (N≥3 neutral transitions) before the next signal is valid. XRPUSDT — live results: 60%+ win rate across 34 loops, LONG/SHORT perfectly symmetric. The price is the shadow. The structure is the signal. #AlgoTrading #Quant #Entropy #MarketMicrostructure #XRPUSDT

#XRPUSDT $XRP SKA Trading Bot v1 — Report (63 loops) ================================================== Date: 2026-03-17 08:18:38 SUMMARY   Total trades: 1586   Winners: 1002 | Losers: 406 | Flat: 178   Win rate: 63.2%   Total PnL: +0.382100   Avg PnL/trade: +0.000241   Best trade: +0.003400   Worst trade: -0.002100 BY SIDE   LONG:  795 trades | PnL=+0.198500 | win_rate=62.8%   SHORT: 791 trades | PnL=+0.183600 | win_rate=63.6% PER LOOP   Loop   1:   9 trades | win= 66.7% | PnL=+0.001600   Loop   2:  24 trades | win= 83.3% | PnL=+0.008600   Loop   3:  22 trades | win= 77.3% | PnL=+0.011000   Loop   4:  19 trades | win= 78.9% | PnL=+0.009600   Loop   5:  28 trades | win= 67.9% | PnL=+0.005900   Loop   6:  30 trades | win= 63.3% | PnL=+0.012200   Loop   7:  29 trades | win= 69.0% | PnL=+0.009500   Loop   8:  30 trades | win= 63.3% | PnL=+0.006200   Loop   9:  31 trades | win= 64.5% | PnL=+0.008500   Loop  10:  26 trades | win= 65.4% | PnL=+0.006900   Loop  11:  28 trades | win= 64.3% | PnL=+0.007500   Loop  12:  18 trades | win= 77.8% | PnL=+0.008800   Loop  13:  19 trades | win= 63.2% | PnL=+0.006200   Loop  14:  26 trades | win= 57.7% | PnL=+0.006800   Loop  15:  21 trades | win= 61.9% | PnL=+0.009100   Loop  16:  27 trades | win= 66.7% | PnL=+0.006000   Loop  17:  28 trades | win= 71.4% | PnL=+0.008400   Loop  18:  30 trades | win= 70.0% | PnL=+0.009300   Loop  19:  28 trades | win= 75.0% | PnL=+0.009900   Loop  20:  23 trades | win= 69.6% | PnL=+0.006400   Loop  21:  26 trades | win= 61.5% | PnL=+0.004300   Loop  22:  26 trades | win= 53.8% | PnL=+0.005100   Loop  23:  28 trades | win= 60.7% | PnL=+0.004800   Loop  24:  32 trades | win= 56.2% | PnL=+0.005400   Loop  25:  30 trades | win= 60.0% | PnL=+0.008000   Loop  26:  24 trades | win= 50.0% | PnL=-0.001600   Loop  27:  25 trades | win= 60.0% | PnL=+0.005500   Loop  28:  22 trades | win= 68.2% | PnL=+0.001000   Loop  29:  25 trades | win= 76.0% | PnL=+0.011100   Loop  30:  25 trades | win= 68.0% | PnL=+0.006000   Loop  31:  22 trades | win= 63.6% | PnL=+0.006400   Loop  32:  23 trades | win= 78.3% | PnL=+0.010600   Loop  33:  29 trades | win= 58.6% | PnL=+0.007100   Loop  34:  26 trades | win= 65.4% | PnL=+0.003800   Loop  35:  25 trades | win= 48.0% | PnL=+0.001900   Loop  36:  24 trades | win= 62.5% | PnL=+0.006000   Loop  37:  23 trades | win= 47.8% | PnL=+0.001600   Loop  38:  31 trades | win= 54.8% | PnL=+0.009200   Loop  39:  32 trades | win= 50.0% | PnL=+0.007400   Loop  40:  26 trades | win= 65.4% | PnL=+0.007600   Loop  41:  31 trades | win= 51.6% | PnL=+0.002400   Loop  42:  20 trades | win= 40.0% | PnL=-0.000400   Loop  43:  29 trades | win= 72.4% | PnL=+0.005900   Loop  44:  19 trades | win= 63.2% | PnL=+0.005900   Loop  45:  22 trades | win= 68.2% | PnL=+0.005800   Loop  46:  25 trades | win= 56.0% | PnL=+0.002000   Loop  47:  24 trades | win= 41.7% | PnL=+0.001200   Loop  48:  20 trades | win= 60.0% | PnL=+0.004600   Loop  49:  22 trades | win= 63.6% | PnL=+0.008200   Loop  50:  20 trades | win= 55.0% | PnL=+0.005700   Loop  51:  31 trades | win= 48.4% | PnL=+0.003400   Loop  52:  33 trades | win= 66.7% | PnL=+0.007200   Loop  53:  28 trades | win= 85.7% | PnL=+0.007900   Loop  54:  21 trades | win= 61.9% | PnL=+0.005900   Loop  55:  26 trades | win= 69.2% | PnL=+0.006700   Loop  56:  24 trades | win= 54.2% | PnL=+0.002300   Loop  57:  20 trades | win= 55.0% | PnL=+0.004600   Loop  58:  16 trades | win= 75.0% | PnL=+0.004400   Loop  59:  30 trades | win= 63.3% | PnL=+0.006000   Loop  60:  22 trades | win= 59.1% | PnL=+0.005100   Loop  61:  26 trades | win= 76.9% | PnL=+0.010000   Loop  62:  31 trades | win= 51.6% | PnL=+0.001400   Loop  63:  26 trades | win= 69.2% | PnL=+0.006300

#XRPUSDT SKA Trading Bot v1 63 loops. 1,586 trades. 63.2% win rate. 61/63 loops profitable. No indicators. No thresholds. No stop loss. The entropy is computed in real time by the SKA learning engine. The signal fires on a topological event — a complete paired regime cycle confirmed in entropy space. Not a price level. Not a threshold. A structural fact. LONG and SHORT are perfectly symmetric: - LONG: 795 trades | +0.1985 | 62.8% win - SHORT: 791 trades | +0.1836 | 63.6% win This is Paired Cycle Trading built on the SKA learning engine. The edge is structural. It does not decay. GitHub github.com/quantiota/SKA-… #AlgoTrading #QuantTrading #CryptoTrading #Binance #EntropyTrading #OpenSource #TradingBot #MarketStructure #SKA

@LePoint Yann Le Cun est un dinosaure car emprisonné malgré lui dans l'ancien paradigme. Il faut des jeunes pour définir une intelligence artificielle basée sur des premiers principes.

Yann Le Cun réunit un commando mondial pour réinventer l’intelligence artificielle depuis la France Par Guillaume Grallet l.lepoint.fr/uDd

@rohanpaul_ai Very dangerous.

A UK survey says that more than 33% of adults are now using AI to support their mental health and well-being. And this is actually something that psychologists are starting to encounter in their work too.

@predict_addict True: Whatever we well understand we express clearly, and words flow with ease. Nicolas Boileau-Despréaux - L'Art poétique

There are no bad math students, only bad math teachers and bad math books. 📚

@shoukointech True

Peter Thiel: Computer Science was a fake field