Activity
Mon
Wed
Fri
Sun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
What is this?
Less
More

Memberships

AI Systems for Human Content

2.5k members • Free

Trading With Mustafa

2.7k members • Free

The AI Advantage

125.4k members • Free

AI Money Lab

83k members • Free

AI Automation Agency Hub

325.5k members • Free

ZeroOne Systems

9.7k members • Free

Trade Algorithm Free

3.1k members • Free

8 contributions to ZeroOne Systems
55000%+ profit trading MNQ using Claude and ChatGPT
I’ve built a TradingView strategy I call AEE — Adaptive Edge Engine with Claude and ChatGPT. The current version on MNQ 5m is showing: - Net P&L: ~$2.76M (initial capital was $5000) - P&L (%): ~55,270% - Profit Factor: ~5.99 - Win Rate: ~75% - Max Drawdown: ~$1.3K (6%) - Trades: ~38K over entire history of MNQ The real breakthrough here wasn’t just “building a profitable strategy.” The key developing the full research and validation workflow using Claude and ChatGPT around the strategy. The workflow now includes: - TradingView strategy development - AI-assisted Pine Script review - Trade telemetry extraction - Session / grade / confidence / trigger analysis - Monte Carlo simulation - Apex-style (Prop Firm) pass/fail survival testing - Red-tail drawdown diagnostics - Daily/session risk analysis - Live execution testing through TradersPost One thing I learned very quickly...a TradingView backtest by itself is not enough. You also need to know: - What happens if you start trading at the worst historical point? - What happens with extra slippage and commissions? - What happens if the edge decays? - What happens if trades cluster badly? - Which sessions actually make the money? - Which signals quietly create drawdown? - Whether your live alerts match the strategy tester behavior? This is not a guarantee of future results, but it has been eye-opening to see how much better the process becomes when AI is used as a research partner instead of just a “write me a strategy” tool. The real value of AI here has been in: 1. Finding failure modes 2. Challenging assumptions 3. Comparing strategy variants 4. Building telemetry tools 5. Stress-testing the equity curve 6. Reducing random trial-and-error I’m now working on turning the Telemetry Analyzer + Monte Carlo Simulator into tools that other traders can use on their own TradingView strategy exports. Curious if anyone else here is using Claude, ChatGPT, or other LLMs this way, i.e., not just to generate code, but to build the full research, validation, and risk-analysis workflow around a strategy.
55000%+ profit trading MNQ using Claude and ChatGPT
1 like • 17h
@Nico Jenkins That’s exactly the kind of workflow I’ve been trying to move toward as well, i.e. not just “ask Claude to build a strategy,” but use it as a research partner across the full process: hypothesis generation, code review, backtest diagnostics, telemetry analysis, and stress testing. And I completely agree with your skepticism. A single-ticker result this high should raise eyebrows, which is why I’ve been spending more time on validation than on the headline P&L number itself. A few important clarifications: 1. This is MNQ, so the result is coming from very high trade frequency over many years, not from a few giant directional calls. 2. The strategy is not just trying to “predict trend.” A big part of the edge appears to come from exploiting short intraday momentum/mean-reversion bursts during specific sessions, especially the 08–10 RTH window. 3. I’m not relying on the TradingView backtest alone. I’ve been exporting trade telemetry into Python and reviewing things like session performance, trigger quality, confidence calibration, rolling-state behavior, Monte Carlo survival, contiguous-start stress, extra slippage/friction assumptions, and Apex-style drawdown pass/fail modeling. Your Amibroker + macro-variable approach sounds really interesting. I’d love to hear more about how you structure the 140-variable review, especially whether you are using Claude more for interpretation, feature selection, regime labeling, or actual rule generation.
Day 1+: CoinPulse, a Crypto Screener that tells you when (or why not) to trade
Hi everyone, I have built CoinPulse, a real-time crypto intelligence terminal for Robinhood traders, based on what I learned from the content for Day 1 and a few other Days after that. Robinhood has 70+ tradable cryptos, but no serious screener. CoinPulse scans them every 90 seconds and answers: * Which cryptos are moving right now? * Is the move real or just a pump & dump? * Is an entry still worth it? * What’s the target, stop, upside, downside, and potential Reward:Risk? * If the system blocks a trade, was that block actually correct? That last part is the real differentiator. CoinPulse tracks every prediction and every blocked signal across 5, 15, 30, and 60-minute windows. The system is configured to send iMessages for qualified setups. It shows whether each filter saved money or caused missed profit. So instead of guessing whether a rule is too strict, the system builds evidence over time. Current focus: * Robinhood crypto only * Long-only * No leverage * No margin * Paper/advisory mode first * Designed to avoid chasing momentum spikes Would love feedback from other traders/builders: is this something you’d use, test, or pay for?
Day 1+: CoinPulse, a Crypto Screener that tells you when (or why not) to trade
0 likes • 13d
@Yosef Yosefi Hi Yosef, You asked, and I delivered. getcoinpulse.com is live. You can see my post on this topic here - https://www.skool.com/zero-one/coinpulse?p=ff499e63 Would love for you to try it out as an early beta user and give me honest feedback, especially on whether the dashboard/signals are useful, what feels confusing, and what would make you want to use it regularly.
0 likes • 13d
@Alistair Smith Hi, I just launched the beta version of CoinPulse:https://getcoinpulse.com I shared the details here:https://www.skool.com/zero-one/coinpulse?p=ff499e63 Would love for you to try it and give me honest feedback. You had noticed the “cohesive solution” angle earlier, so I’d be especially interested in whether the beta now feels clear and useful from a trader’s perspective.
CoinPulse Beta Launch
Several members asked me if I had a URL for the crypto tool I had shared earlier this week in the community, so I pushed myself to turn the localhost version into a real web beta. It’s called CoinPulse:https://getcoinpulse.com It scans about 75 Robinhood-tradable cryptos every minute and tries to answer: What’s moving, why is it moving, and is it still worth reviewing — or am I already late? Current beta features: - Live crypto momentum dashboard - Buy / Watch / Wait / Avoid signal logic - Signal accuracy + target-hit tracking - “Why is it moving today?” context - Do Not Chase / blocker detection - Simple + Pro dashboard modes - Trading style presets - Optional SMS alerts - PulseBot, an AI chatbot that gives context-sensitive help and explains what you’re seeing No auto-trading and no brokerage connection (yet)... and no financial advice — just a research tool to help review crypto momentum setups more clearly.... made entirely based on the prompts shared by Lewis in different videos and inspired by the ideas of daily project updates posted by numerous members. Thank you Lewis and everyone! I’d love feedback from active crypto traders. What feels useful, what feels confusing, and what would make you actually use this daily? Please let me know. Thank you in advance to everyone for their inputs.
CoinPulse Beta Launch
0 likes • 13d
@Justin Valentine That’s really helpful, and honestly that’s exactly the direction I’m starting to think about. I started with the Robinhood-tradeable universe mainly because it gave me a clean retail-focused set of coins to compare against each other, not because CoinPulse is actually integrated with Robinhood. But based on the feedback I’m getting, adding a broader Binance/Kraken/Coinbase-style universe makes a lot of sense. The product probably gets more useful when it can rank a larger pool of coins and then let users filter by where they actually trade. And yes, the data-cost side matters a lot. If Binance market data can support a broader universe without creating a huge API cost problem, that could be a big plus. Really appreciate the “I would buy it” comment too. That’s exactly the kind of feedback I need during beta.
1 like • 13d
@Justin Valentine Agreed on both points - using a broader crypto market and perhaps using CoinGecko API. The CoinGecko API would be useful, especially for metadata/context, but rate limits and pricing tiers can become a real scaling issue if the app starts polling a lot of symbols frequently. I’m thinking the better architecture may be to use exchange-native data where possible for real-time price data, then use CoinGecko more selectively for metadata, links, categories, and context. Appreciate the feedback again. Thank you so much.
Options and Futures Trading Mentor
Thanks for the add. I have been trading stocks since 1990, options and futures since 2005. I was introduced to "probability-based" options trading in 2011. I wrote and published a book on Amazon (Introduction to Probability-Based Options Trading). In discovering the Markov Method on your videos, I have found Markov to be a great adjunct to what I coach others on. I using Claude Code to include Markov Method biases in my morning "Futures Briefing" I distribute to members. If you would like a copy, I am happy to add you to the distribution list. Thanks Lewis for great information.
2 likes • 17d
@Renier Lemmens, I went through the exact same journey. Markov is a great descriptor but has zero predictive value on its own. I hit that wall early, so decided to dive deeper. I ended up building a multi-model regime indicator that runs 6-7 models simultaneously with consensus voting, such as Markov, HMM (trained in Python on daily data), plus Hurst (for trend), volatility, oscillator, Bernoulli (for noise/entropy), and persistence models. Each one captures something different that Markov alone misses. The key finding: no single model predicts direction, but the consensus works as a trade filter. When 5/5 models agree on Bull, the environment is measurably different from 2/5 or 3/5. I don't use this to predict buy/sell, instead I use it to decide whether to trade or not. The win rate improvement is modest (~3-5%), but the drawdown reduction is real because you stop trading during the uncertain/choppy/worst regimes. The attached screenshots show: - Status panel — Entropy (model disagreement), volatility regime, Hurst persistence, and consensus (Bull 5/5 here) - MTF grid — same models across 1m through 4H. Green triangles lining up = genuine multi-timeframe alignment - Model dashboard — each model's regime call, bull/bear probability, persistence, and confidence @Claudio Damian Catalani, I tested multiple horizons. Regime doesn't predict t+1 price, but filtering entries by consensus (only trade when 4/5+ agree) reduces drawdown meaningfully. Built this for my own MNQ day trading. Happy to discuss the methodology if anyone's interested.
ChatGPT / Auto trade / Tradeovate / Prop Firms
In regard to auto trading, or placing trades how can we get Chat GPT - Open AI to do this? Would want to use the one-shot prompt for Claude Code and apply it to Tradeovate, which is the broker I would use on Trading View. Is anyone else trying to do this for Prop Firms?
2 likes • 25d
Hi Shane, yes, I am using the exact same setup. TradingView to TradersPost.io to Tradovate to Apex Trading using automated trading strategy in TradingView. However, I am not configuring ChatGPT to trade for me. I am using ChatGPT/Claude only to create the Trading Strategy, and the automation of the strategy is simply through TradersPost.io.
0 likes • 18d
I have just been working with Claude iteratively. Nothing to it. I don't know coding, so I just rely on the results I am seeing, and tell Claude to fix things when they don't look right.
1-8 of 8
Anuj Saxena
3
44points to level up
@anuj-saxena-6624
https://www.linkedin.com/in/anujsaxena2/

Active 2h ago
Joined May 12, 2026
Powered by