Blog
AI trading signals, researched in the open
How multi-agent AI councils analyze markets, how to backtest crypto strategies without risking capital, and what our own council data says about which models trade best. RSS.
Prompt Engineering for Market Analysis: Lessons from Running Agents in Production
Getting reliable market analysis out of an LLM is mostly prompt discipline: feed data as narrative rather than raw tables, enforce a strict response format, give each role a sharp mandate, and treat every prompt edit as a strategy change worth versioning.
2026-07-15
AI trading signalsFree AI Trading Signals: What You Get and What to Verify
Free AI signals range from genuine free tiers to loss-leaders for something else. The test is what you can see and verify without paying: the reasoning, the full track record, and whether your data or funds are the real price.
2026-07-12
Technical analysisThe Indicator Stack an AI Analyst Reads: RSI, ADX, DMI, and ATR
RSI measures momentum extremes, ADX measures trend strength, DMI measures trend direction, and ATR measures volatility. Together they answer four different questions — and reading them as a stack, not in isolation, is what separates signal from noise.
2026-07-11
Model comparisonsThe Cheapest Council That Works: Cost-per-Signal Across Model Providers
Model cost varies wildly, and not every council seat needs the flagship. The smart build puts cheap, capable models on the summarizing tiers and spends on the seat that makes the final call — cutting cost-per-signal without gutting quality.
2026-07-11
Market dataWhat Is an OHLC Candle, and Why Timeframe Changes Everything
An OHLC candle summarizes price over an interval as Open, High, Low, Close. The interval you pick isn't cosmetic: a 15-minute and a daily council read genuinely different markets, because every indicator is computed from the candles you chose.
2026-07-11
Run a council yourself
Reading about it only goes so far. Spin up an AI council on a live ticker — three free, no card required.