For self-directed investors

Build and stress-test a portfolio based on what you actually believe.

Write your investment thesis. We score every candidate against a fundamental framework: archetypes, valuation, regime fit, and fragility. Then we build a portfolio with the reasoning to back it.

Start with your thesis See how it works

Thesis Workspace

live context

1) Thesis layer

Thesis vector

AI capex cycle across compute, software, cyber, and power infrastructure.

long horizon thematic risk-aware global universe

2) Stock scoring layer

Candidate scorecard

Quality

Value

Fragility

MIS

Archetype

Eagle

Confidence

High

3) Portfolio layer

Portfolio construction

Core Infrastructure Contrarian

Conviction62% Optionality38%

Re-rank on live data and reasoning checks.

Signal Field

adaptive intelligence

Universe 9.5k Regime-aware Archetype model Live repricing

Volatility lens

Dynamic

Thesis coherence

High

Bias defense

Active

Two entry points

One tool. Two ways in.

Have a thesis?

Write it. We find the stocks that fit and stress-test the basket against your risk and horizon.

Start with your thesis ->

Have a portfolio?

Import it. We test whether it actually holds up under your thesis and current market regime.

Import your portfolio ->

Most investors do not need more data. They need a better process.

How it works

Three steps. Honest and specific.

Write your thesis

Not a form. A conversation. Tell us what you believe about a sector, trend, or macro shift. The thinking partner helps you sharpen it into a testable thesis.

The engine scores

Every candidate runs through deterministic scoring: business quality, valuation attractiveness, regime fit, and fragility. Same inputs, same outputs.

Your portfolio, with reasoning

Get 10-15 scored names with a report explaining why the basket fits your thesis, horizon, and environment. Then challenge it, swap names, and rerun.

A strategy you can follow beats one you abandon.

Thinking partner

Not a chatbot. A partner with full context.

It knows your thesis, your portfolio, each score, and the reasoning behind every pick. Ask why a stock was included, request alternatives, or challenge assumptions directly.

It pushes back when reasoning is weak and explains scores in plain language when you want that. General AI tools cannot do this because they do not hold your full investment context.

Structure over noise. Conviction over impulse.

Thinking Partner full thesis context

You flagged PANW as expensive. Do you want lower multiple alternatives in the same thesis layer?

Yes. Keep cybersecurity exposure but reduce valuation risk.

Understood. Replacing with CHKP improves valuation and fragility while preserving thesis fit.

What investors experience

First: clear fit. Then: deeper insight.

Recognition in practice

A thesis-matched basket that already feels right.

You get a diverse, coherent exposure set: core names, supporting layers, and selective contrarian optionality. Not random picks. A structured basket that reflects the thesis.

Direct plays + adjacent enablers in one portfolio view
Contrarian picks framed as optional, not forced
Coverage that is broad enough to reduce single-point risk

Useful surprise in practice

A counterintuitive idea that earns its place.

The engine surfaces opportunities that are easy to miss at first glance, then stress-tests them with scoring and reasoning so you can separate signal from novelty.

Find non-obvious names with measurable thesis relevance
Use valuation, fragility, and regime context to validate
Challenge confirmation bias before capital is allocated

Series investors

Built with the people who already do the work.

Placeholder section for real investor interviews. Layout and style are production-ready; copy and photos can be swapped once validated testimonials come in.

Octavian M.

Operator-investor · Tech infra focus

“It surfaced most of the names I already considered serious, then added two I had ignored. The key difference was seeing the reasoning and fragility checks side by side.”

Recognition + disciplined extension

Lea K.

Long-horizon allocator · Macro + quality

“I’m not looking for trade calls. I want a process I can defend. Thesis gave me a portfolio I could challenge, edit, and rerun without losing the logic.”

Process confidence over noise

Daniel R.

Independent PM · Fundamental + contrarian

“The useful surprise wasn’t a random ticker. It was a counterintuitive candidate backed by valuation, regime context, and a clear bias check before I touched allocation.”

Counterintuitive, but testable

The framework

A process you can audit, not just trust.

Step 1 - Inputs

Investor context + thesis intent

We start with your profile, horizon, risk tolerance, and explicit thesis statement. That defines the decision boundary before any ranking begins.

Step 2 - Deterministic core

Institutional-grade scoring engine

Each candidate is evaluated across long-horizon financial history, valuation structure, business resilience, and regime sensitivity. Same inputs, same outputs, full repeatability.

Step 3 - Contextual reasoning

Cross-reference, challenge, refine

The system reasons over scored sets, tests for weak assumptions, and highlights non-obvious opportunities without dropping the quantitative discipline.

Step 4 - Portfolio output

Ranked basket with explicit rationale

You get a thesis-aligned portfolio with conviction tiers, contrarian optionality, and transparent reasoning so you can edit, defend, and rerun it.

See a scored portfolio example ->

Start now

Build from conviction. Test with rigor.

Your thesis keeps you invested. The market keeps you sharp.

Start with your thesis Import your portfolio