About / Field Manual
AI Memos
How AI memos, targets, scenarios, and summaries are generated.
AI Memos combine structured signals with model-guided narrative synthesis.
AI Memos are a research memo layer that combines structured signals (price, factors, fundamentals, risk) with model-guided narrative synthesis. It produces scenario-based price targets (bull/base/bear), a confidence estimate, and a compact thesis. The goal is speed and consistency, not certainty.
Scenarios & target ideas
Three forward-looking paths based on current data and model estimates.
Bull Case
Upside scenario if catalysts materialize and conditions improve.
Base Case
Our central scenario given current signals and typical conditions.
Bear Case
Downside scenario if risks skew worse than expected.
Inputs
Structured signals that feed the AI memo engine.
Price Action
Trend, momentum, volatility, volume, and liquidity conditions.
Factors
Style, quality, value, sentiment, and cross-sectional signals.
Fundamentals
Financials, estimates, ratios, revisions, and valuation context.
Catalysts
Earnings, guidance, product, regulatory, and event triggers.
Risk
Drawdowns, leverage, correlation, tail risk, and regime context.
Inputs are sourced from point-in-time fundamentals and fresh market context. The system anchors scenarios to a real observed price at the time of analysis, then enriches with filings, news, and externally sourced analyst targets.
Outputs
What the AI memo system produces.
Compact Thesis
A concise statement of what we think, why it matters, and the key drivers.
Scenarios
Bull, Base, Bear paths with probabilities and qualitative rationales.
Price Targets
Target ranges for each scenario with assumptions and time horizon.
Summaries
Investor-ready takeaways pulled from the memo for quick scanning.
Output is a memo snapshot tied to a specific analysis date. The UI presents both a compact summary and a full-detail memo view. Memos are latest entries across covered companies; Memo Summary is the condensed stock-page view; AI Memo is the full write-up.
Quality & limits
AI memos are research tools, not guarantees. They can reflect model error, outdated or incomplete data, and unforeseen events. Always do your own research and consider multiple perspectives.
- Use as decision support, not as advice
- Model outputs can change as data changes
- Past performance is not predictive of future results
Refresh cadence
AI memos run on a focused universe and refresh bi-weekly or shortly after earnings. If a memo is missing, the UI falls back to the latest available snapshot.