BTCBuzz: A Self-Auditing Bitcoin Signal Bot Tested with Jev
BTCBuzz is an educational Bitcoin signal bot built by Cloud First Consulting. It paper trades only, with no real money, and nothing in the article is financial advice. Every night at 00:40 UTC it scores yesterday's call, checks its validation gates, fetches price, derivatives, on-chain, macro and news data, runs a RandomForest, XGBoost, LightGBM and PyTorch LSTM ensemble, and posts a BUY, SELL or HOLD report to Telegram. Track record so far: a walk-forward backtest of 718 simulated trades across 9 folds, where holding each signal to its 30-day horizon returned +2.82% per simulated trade at an out-of-sample Sharpe ratio of 0.60, against -0.46% at -0.36 for the exit ladder the bot first published and retired on 5 September 2026. The live paper book had closed 4 simulated trades by 17 September 2026, all 4 in profit, for +$34.19 on a simulated $10,000 account, and the bot's own report tags any rate based on fewer than 20 results as not statistically meaningful. The bot's weak spot was three rules that read news by counting keywords. On 19 September 2026 we ran Jev, the System One model from TypeSafe.ai (jev-1.13.0, $0.042 per million input tokens, output free), beside the keyword scorer on 1,499 archived headlines in shadow mode. The keyword scorer flagged 273 adverse events, Jev flagged 282, and both flagged the same 99. Jev judged 106 of the 273 keyword flags (39%) not material to crypto. Scoring the archive took 1,451 calls and cost $0.0686 with 0 failures, median latency 711 ms. Jev cannot change a signal until it passes a blind-labelled test of 150 headlines whose pass mark was written first. A companion article, What Is Jev, explains the model itself. Cloud First Consulting has no commercial or other relationship with TypeSafe.ai.
Frequently Asked Questions
- Is BTCBuzz financial or investment advice?
- No. BTCBuzz is a research and educational project built by Cloud First Consulting to study how machine learning, validation gates and AI agents behave on a hard, noisy problem. It does not give financial, investment or trading advice. All of its trading is simulated paper trading with no real money, every figure in this article comes from simulation or from software measurements, and nothing here is a recommendation to buy, sell or hold Bitcoin or any other asset. Cloud First Consulting is not authorised or regulated by the Financial Conduct Authority to give investment advice.
- What is BTCBuzz?
- BTCBuzz is a Bitcoin signal bot written in Python. Once a day, after the 00:00 UTC close, it scores yesterday's prediction, checks its own validation gates, collects price, derivatives, on-chain, macro and news data, runs an ensemble of RandomForest, XGBoost, LightGBM and a PyTorch LSTM, and writes a BUY, SELL or HOLD report with its reasoning. It paper trades its own signals and posts the report to a Telegram group. A background research loop proposes changes to the strategy, and the validation gates reject the ones that only look good in a backtest.
- How has BTCBuzz performed so far?
- The evidence of record is a walk-forward backtest of 718 simulated trades across 9 folds, where holding each signal to its 30-day forecast horizon returned +2.82% per simulated trade at an out-of-sample Sharpe ratio of 0.60. The live paper book is four weeks old: by 17 September 2026 it had closed 4 simulated trades, all 4 in profit, for +$34.19 on a simulated $10,000 account. The bot's own report labels any rate based on fewer than 20 results as not statistically meaningful, and that applies to the paper book. These are simulated results and say nothing about future returns.
- What is Jev from TypeSafe.ai?
- Jev is a model from TypeSafe.ai that the company calls a System One model. You send it a piece of text and a set of typed questions, and it returns structured answers: a probability for a yes or no question, a probability for each option in a multiple choice question, or a probability for each level of an ordered scale. It does not write prose. At the time of writing the listed price was $0.042 per million input tokens with no charge for output, and the model version we pinned was jev-1.13.0. Our companion article, What Is Jev, explains the model in full.
- How much does Jev cost to run in BTCBuzz?
- Each headline costs about 1,120 input tokens for our set of 11 questions. At roughly 50 headlines a day that is about $0.0024 a day, or less than $1 a year. Scoring our whole 28-day archive of 1,499 headlines (1,508 before duplicates were removed) took 1,451 API calls and 1,632,955 tokens and cost $0.0686, with no failed calls.
- Does Jev make BTCBuzz more accurate?
- We do not know yet, and we have not claimed it does. Across 1,499 archived headlines Jev and our keyword scorer flagged almost the same number of adverse events (282 and 273) but agreed on only 99 of them. Nobody has yet judged which reading was right. We wrote the pass and fail criteria before running anything: 150 headlines labelled blind by a person, then per-class F1 for both scorers with confidence intervals. Until that is done, Jev runs in shadow mode and cannot change a signal.
- What does shadow mode mean?
- The new component runs on live data next to the old one, and its output is written to a log beside the old output, but only the old output is used. In BTCBuzz nothing Jev returns can change a signal, a position size or a published number.
Our Services
Contact Cloud First Consulting
Email: info@cloudfirstconsulting.com
Location: London, United Kingdom
Hours: Monday-Friday, 9:00 AM - 6:00 PM GMT
Book a Free 30-min Discovery Call