Layer 1 -- The Panel (six specialists)
HELIOS, NEXUS, ARGOS, VEGA, ATHENA, and PSYCHE each read the week through a single domain lens with explicit boundaries designed to reduce cross-contamination. HELIOS covers rates, liquidity, and financial conditions. NEXUS covers innovation cycles and risk appetite. ARGOS covers geopolitics, energy, and commodities. VEGA covers emerging markets, FX, and capital flows. ATHENA covers fundamentals, quality, and valuation. PSYCHE covers positioning, sentiment, and reflexivity.
Each week, the full scenario is evaluated against the curated ~40 US-listed ETF universe; the process then selects the ETFs best positioned for the current context — forming that edition's theoretical portfolio for deep, focused analysis. The Panel's job is not to agree -- it is to give each domain its own honest read before any synthesis happens. That separation is what makes the eventual disagreements informative rather than an artifact of shared assumptions.
Layer 2 -- Five independent SENIOR verdicts
SENIOR instances are architecturally separate pipelines -- same inputs in, five fully independent conclusions out. The five SENIOR instances run on different model architectures: Google Gemini, OpenAI GPT, Anthropic Claude, DeepSeek, and xAI Grok. This is not five copies of one model with different names; it is five genuinely distinct deliberation paths evaluating the same evidence package.
The investor-relevant output from the SENIOR layer is not simply "what they said individually" -- it is whether they agree, where they split, and how large the disagreement is. A four-to-one consensus tells you something different than a two-to-three split, even if the majority recommendation appears the same. That second-order information about the shape of the verdicts is as important as the verdicts themselves.
Layer 3 -- PRIME synthesis
PRIME applies a fixed protocol to the five SENIOR outputs every week: consensus degree, divergence mapping, evidence-quality flags, and systemic bias risk assessment. The output is a week-to-week comparable synthesis -- not a free-form editorial summary that changes character based on who wrote it or what the market did.
The fixed-protocol nature of PRIME is what enables auditability over time. Because the same protocol runs every week, you can look at this Friday's PRIME and compare it directly to six months ago: which dimensions showed the most tension? Which SENIOR verdicts were in the minority but turned out to be right? Where did evidence quality warnings appear before a scenario shifted? That retrospective view is how process-driven investors build a calibrated sense of which signals matter.
Why three layers instead of one
Single-layer products -- one AI answer, one analyst summary, one consolidated recommendation -- optimize for comfort and decisiveness. They compress uncertainty into a clean output, which feels helpful but trades away the information contained in the disagreement. You cannot tell from a single verdict which dimension drove the conclusion, or how contested the scenario was, or where the evidence was thin.
The three-layer stack trades that false sense of certainty for structured visibility. You get more cognitive work upfront, but you get something much more valuable: a map of where the scenario is solid and where it is contested, week after week, on a consistent protocol. This is still general information, not advice -- but it is information you can actually think with.
How this connects to your Friday dossier
Every Friday, the dossier you receive contains: the full Panel output (six specialist analyses), all five SENIOR verdicts, the PRIME synthesis with consensus and divergence mapping, and a companion closure document focused on portfolio-perspective analysis -- highlighting relevant entries and exits for that week and long-term reads for positions you plan to hold.
The closure document is not portfolio management (that decision is yours, and involves your personal suitability, tax situation, and risk tolerance). It is portfolio-relevant analysis -- a step beyond generic research that connects the week's scenario to the specific instruments you might be watching. Free subscribers access the full dossier 7 days after publication; Pro subscribers receive it same-day plus Monday-Thursday briefings (two per day) keeping the thesis updated through the week.
Why the protocol matters over time
Because PRIME applies the same protocol every week, you accumulate a structured record that is genuinely comparable over time. You can look back six months and see which lenses flagged what first, which SENIOR verdicts were in the minority and proved correct, where PRIME's evidence quality warnings appeared before a significant scenario shift.
That auditability is how you build genuine trust in a process -- not blind faith in an algorithm or an individual. It also lets you calibrate: over time, you learn which dimensions of the analysis have historically been most informative for your specific ETF interests, and you can weight your reading accordingly. That calibration is an asset that compounds, independent of whether markets go up or down.