How Does Acubic Work? A Complete Guide

Key takeaways

• Acubic is an AI portfolio builder with optional broker automation for eToro and Trading 212.
• You describe your amount, risk tolerance and limits in a short guided conversation, and those answers become hard constraints on the portfolio.
• Weights are set by a stated method, such as hierarchical risk parity or a rule-based weighting, chosen when you build the strategy, not by return forecasts alone.
• You review the proposed holdings and weights before anything reaches a broker, and one-click approval is the default mode.
• Acubic does not promise returns and does not try to predict short-term market moves.

Before anyone connects a brokerage account to an AI tool, they ask a simple question: how does this actually work? It is a fair question, and a financial product should answer it plainly rather than hide behind the word "AI". This guide walks through what Acubic does end to end, from the moment you describe what you want to the point where a portfolio is built and, if you choose, kept aligned at your broker over time.

What Acubic is, in one sentence

Acubic is an AI portfolio builder paired with broker automation. You describe your goals and constraints, Acubic constructs a diversified portfolio using quantitative methods, and you can optionally connect a supported broker so that the portfolio is placed and rebalanced for you. It is a tool for constructing and maintaining a portfolio with discipline, not a system that promises returns or predicts short-term market moves.

Step one: you describe what you want

The process starts with a short guided conversation rather than a blank spreadsheet. You tell Acubic the shape of the portfolio you are after: how much you are investing, your risk tolerance, and any preferences or limits you want respected. Those inputs become explicit constraints, not vague suggestions, which is what lets the rest of the process stay quantitative and repeatable.

This matters because most portfolio mistakes happen at the intake stage, when goals are never written down as constraints. We cover the common versions of this in our write-up on top portfolio construction mistakes.

Step two: the AI builds a diversified portfolio

Acubic then constructs a portfolio using established quantitative techniques rather than a hunch about which stock will go up. The engine works from the risk relationships between holdings, not from short-term price forecasts, and it applies your constraints as hard limits on the result. The goal is a portfolio whose risk is deliberately allocated, rather than one that looks balanced on a fact sheet but concentrates risk in a few correlated names.

If you want the detail behind this, we have written it up separately. Our piece on quantitative portfolio construction that holds up explains the method, and how to optimize a portfolio that fits risk covers how constraints shape the final allocation.

Acubic does not preselect a construction method: you choose one when you build the strategy, from an optimiser such as hierarchical risk parity, mean-variance or CVaR, or a rule-based weighting such as equal weighting. For Beginner and Intermediate profiles the menus recommend rule-based weightings; Advanced users see the optimisers side by side with no recommendation. Hierarchical risk parity, for example, does not bet on which names will outperform: it groups holdings by how their returns move together and spreads risk across those groups, which stops the portfolio from quietly concentrating in a handful of correlated positions. Plain mean-variance optimization is very sensitive to small errors in return estimates and tends to produce concentrated, unstable portfolios, which is why it is offered alongside methods that do not depend on return forecasts.

The major problem with MV optimization is its tendency to maximize the effects of errors in the input assumptions.

That observation, more than three decades old, is why Acubic offers methods that do not depend on return forecasts. For the general mechanics behind tools like this one, the explainer on what happens inside an AI portfolio generator covers inputs, data, ranking and weighting step by step.

Step three: you review before anything is placed

A portfolio suggestion is not an order. You see the proposed holdings and weights first, and nothing reaches a broker until you decide to act on it. This human-in-the-loop step is deliberate: it keeps you in control of your own money and keeps Acubic in the role of a tool that does the construction work, not one that trades without your say-so.

Step four: optional broker automation

If you want Acubic to place the portfolio for you, you can connect a supported broker. Acubic currently works with eToro and Trading 212. Broker credentials are encrypted at rest, and the connection is scoped to the account you choose to link.

Acubic offers two automation modes. In one-click mode, which is the default, the scheduler prepares each rebalance and flags it for your approval, and nothing is placed until you approve it. In full-auto mode, the scheduler carries out the scheduled rebalance for you. The normal path therefore keeps a human approval step in the loop, and you opt in to full automation only if you want it.

Once connected, Acubic can keep the portfolio aligned with its target over time. Portfolios drift as prices move, so periodic rebalancing brings weights back toward the plan. The U.S. Securities and Exchange Commission defines rebalancing as "bringing your portfolio back to your original asset allocation mix" (SEC, Beginners' Guide to Asset Allocation), and FINRA's report on digital advice tools puts it the same way: "Rebalancing an investment portfolio is necessary to maintain a target asset allocation over time" (FINRA, Report on Digital Investment Advice, 2016). We explain the mechanics, including timing and thresholds, in our portfolio rebalancing strategy model guide, and why rebalancing and optimization are different jobs in portfolio optimization vs rebalancing.

The four steps at a glance

Step	What you do	What Acubic does
1. Describe	Answer a short guided conversation: amount, risk tolerance, limits	Turns the answers into explicit constraints
2. Build	Nothing	Screens and ranks assets, then sets weights with the method you chose
3. Review	Inspect the proposed holdings and weights	Places nothing until you decide to act
4. Automate (optional)	Connect eToro or Trading 212 and pick a mode	Prepares rebalances on your schedule; one-click approval by default. Full-auto needs a Trading 212 account or an eToro sub-account

What Acubic does not do

Being clear about the limits is part of being trustworthy. Acubic does not promise a return, does not guarantee that any portfolio will beat a benchmark, and does not try to time the market with short-term predictions. It builds and maintains a portfolio according to a stated method and your constraints. Judging the output on the right terms means looking at risk-adjusted measures rather than headline returns alone, which we cover in risk adjusted return analysis.

Where the AI actually helps

The useful role for AI here is not prediction. It is turning a plain-language goal into a structured set of constraints, doing the construction work quickly and consistently, and keeping the portfolio aligned without you having to run the process by hand every few weeks. That is a workflow improvement with oversight, and we set out the honest version of it in AI for portfolio management that adds value.

Getting started

The fastest way to understand Acubic is to build a portfolio and look at the result before connecting anything. You can start with the AI portfolio builder, read the method in more depth in our guides, or learn who is behind the product on the about page. Build first, review the portfolio, and only connect a broker when you are comfortable with what you see.

When the first proposal arrives, the guide on how to read the proposal explains what it contains, what the weights mean, and what it does not tell you.

The short version is that the quiz sets the risk profile, the optimizer builds the portfolio, and the broker connection keeps it there. Which broker you connect changes how the last step behaves: see Acubic with eToro and Acubic with Trading 212 for the specifics of each, and the pricing page for what the Free and Pro tiers include.

Sources

1. The Markowitz Optimization Enigma: Is 'Optimized' Optimal? (Financial Analysts Journal, vol. 45, no. 1), Richard O. Michaud, 1989, accessed 12 October 2026.
2. Beginners' Guide to Asset Allocation, Diversification, and Rebalancing, U.S. Securities and Exchange Commission, 2009, accessed 12 October 2026.
3. Report on Digital Investment Advice, FINRA, March 2016, accessed 12 October 2026.