Skip to content

Cookies are off until you say otherwise

Non-essential cookies stay off unless you accept them, and rejecting them leaves the site fully usable. What runs either way is only what the site cannot work without: the anti-abuse check on our forms, the record of the choice you make here, and the access cookie on internal routes. Site analytics is cookieless and loads only after you accept. Cookie policy

Quant24
Start in Sandbox — free

Team

Two founders and the rule the company was built around. Biographies and photographs are not published yet, and the placeholders say so.

Quant24 is two people: Benjamin Caldera Nieny, CEO and Co-Founder, and Xavier Liberona, CTO and Co-Founder.

That is a small company building something with a large surface area, and it changes what we can honestly promise. It is also why the architecture puts the safeguards in code rather than in a process a bigger team would have to police: a risk manager that holds a veto, an auditor that runs a deterministic checklist, and a rule that no number reaches a page without saying where it came from. Those hold whether or not anyone is watching, which is the only kind of rule worth writing at this size.

Biographies and photographs are not published yet. The placeholders on the story page are deliberate — we would rather show a gap than fill it with a paragraph nobody has approved.

The story of how the company started, and the observation it started from, is below. For anything that needs a person, including press, write to b.caldera@quant24.io.

The founders

  • Photograph not published yet.

    Benjamin Caldera Nieny

    CEO and Co-Founder

    Biography not published yet.

  • Photograph not published yet.

    Xavier Liberona

    CTO and Co-Founder

    Biography not published yet.

How Quant24 started

Quant24 began with an uncomfortable observation. The tools sold to retail futures traders are very good at showing what a strategy would have done and very bad at telling you whether that number means anything. Backtests are easy. Backtests that survive data you never looked at are not, and almost nothing on the market makes the second one the default.

The two of us came at that from opposite ends. One had spent enough time in front of a chart to know how a plausible edge dissolves on contact with slippage and hesitation. The other had spent enough time building systems to know that the only defence is structural: if the honest path is also the easy path, people take it, and if it is not, no amount of good intention holds.

So the rule we started from is the one the product still enforces. The language model decides what to test and explains the result; the code produces every number. An agent cannot report a metric it never computed, an auditor reviews the process instead of the outcome, and a strategy earns capital one level at a time as evidence accumulates — or gives it back.

That rule costs us things. It makes the product slower to demonstrate than a screenshot of an equity curve. It forces us to publish statistics that argue against our own category. It means a page sometimes reads "not available in this build" where a competitor would show a number.

We think that is the trade. A trader who has been sold certainty once can smell it the second time, and the only durable pitch to that person is an audit trail.

We are two people. Everything above is what we are building toward; what is finished and verified, and what is still a mock, is on the changelog. It records no dates for things that have not happened, so there is no roadmap on it and none anywhere else on this site.