About Early Signal
Quantitative AI supply chain investment research, built in the open.
What We Do
Early Signal maps the AI ecosystem across a 10-layer supply chain taxonomy — from raw materials and semiconductor manufacturing through foundation models and AI-enabled enterprise transformation. We generate systematic investment signals by blending daily, weekly, and quarterly data sources, then construct model portfolios with built-in risk management.
Why the Supply Chain View
Most AI investment products track a flat basket of "AI stocks." We think that misses the point. AI isn't one industry — it's a supply chain with distinct layers, bottleneck dynamics, and signal propagation patterns. A surge in hyperscaler CapEx affects chip designers, foundries, and equipment makers differently and on different timescales. Our taxonomy captures those relationships so signals can flow along real commercial links.
How It Works
- Universe curation — 65+ AI supply chain tickers plus a HALO basket of physical-asset beneficiaries, each mapped to primary and secondary layers with upstream/downstream edges.
- Multi-timeframe signals — Daily momentum, options flow, and sentiment. Weekly analyst revisions and sector rotation. Quarterly fundamentals, CapEx, and market share. Blended with layer-specific weights.
- Portfolio construction — Core/satellite/tactical sleeves with position sizing from Fractional Kelly, volatility adjustment, risk parity, and signal proportional methods.
- Risk management — Drawdown-aware regime system that scales exposure from 100% (normal) down to 30% (crisis). Stress testing across macro scenarios. Backtesting against historical data.
What You Get
- Morning Brief — Daily email digest with signal changes, urgency rankings, and risk alerts.
- Dashboard — Real-time signal grid, portfolio view, and bottleneck monitor.
- Weekly Reports — AI-narrated market recaps with top buys, sells, and regime analysis.
- Backtester & Stress Tester — Run historical or scenario analysis against the full universe.
- Simulator — Paper-trade with model portfolio recommendations before committing capital.
Built With Transparency
Every signal shows its component breakdown (daily, weekly, quarterly). Every recommendation shows conviction, target weight, thesis, and key risk. Data provenance labels tell you which inputs are live market data vs. synthetic or estimated. When data quality is imperfect, we say so — the methodology page explains exactly how.
How To Evaluate Us
Finance products should be judged on evidence, not slogans. That is why Early Signal publishes a sample report, a public model portfolio, and a detailed methodology rather than asking you to trust a black box. If you want to pressure-test the work, start there.
What Keeps The Product Accountable
- Public artifacts — methodology, sample report, and model portfolio are visible without a paid plan.
- Clear labeling — delayed snapshots, synthetic inputs, and paper-traded results are all explicitly marked.
- Known limits — this is a research and decision-support tool, not a brokerage and not personalized financial advice.
- Open feedback channel — users can challenge the product directly at [email protected].
What We Will Not Pretend
Early Signal does not promise certainty, and it does not hide uncertainty behind vague AI language. The product can be wrong. Signals can fail. Regimes can shift faster than models adapt. The goal is not perfect prediction — it is a more disciplined way to map the AI supply chain, rank opportunity, and show the reasoning behind each view.
Who This Is For
Self-directed investors, RIAs, and analysts who want a systematic, quantitative lens on the AI supply chain — not tips, not hype, just data-driven signals with full transparency into the methodology.
Contact
Questions, feedback, or a challenge to the methodology? Reach us at [email protected].