Here is a strange fact about banking: some of the highest-stakes decisions in finance are taught with slideshows.
Someone hands a promising analyst a slide deck about credit scoring, a case study or two, and then, eventually, a real loan book with real money and real consequences. If the analyst freezes, or guesses wrong about how a portfolio behaves when the economy turns, the tuition gets paid in write-offs. This is a bad way to learn, and everyone in banking sort of knows it, and mostly they do it anyway because the alternative - letting people practice on live portfolios - is worse.
BankersLab's entire premise is that there is a third option, and that pilots figured it out decades ago. You don't learn to fly by crashing planes. You learn in a simulator, where the stall is real, the recovery is real, the muscle memory is real, and the airplane is imaginary. BankersLab builds that simulator for lenders. Its own word for the product is "flight simulator," and the tagline - "Real-World Data, Fake Risks" - is basically the whole pitch compressed into four words.
The company was founded in 2012 by Michelle Katics, a risk manager, mathematician and economist who got tired of watching training that didn't stick, and Kurt Gingher, who built the simulation algorithms underneath it. The insight was less about banking than about how people learn: you retain what you do, and you especially retain what you do when there's a scoreboard and a team and something at stake that isn't your job. So BankersLab wrapped credit strategy, scorecard management and collections inside a game, gave participants a synthetic portfolio, and let them run it through good times and bad.
The result is the sort of thing that sounds gimmicky until you notice how well it travels. Banks, credit unions, lenders, credit bureaus and training institutes from more than 30 countries - some counts say 45 - have put staff through BankersLab workshops. More than 5,000 professionals have been trained. The workshops run live, over virtual classroom or in person, with expert faculty and the gamification doing the quiet work of making people care about scorecard tracking at 3pm on a Tuesday.
What makes this genuinely interesting, rather than just a nicely produced corporate-training story, is where the simulation engine goes next. If you have spent fifteen years modeling how loan portfolios behave - how delinquency rolls, how a score cutoff trades risk against growth, how a collections strategy plays out across a book - you have built something more general than a teaching tool. You have built a way to test decisions before making them. And that is the same thing lenders actually want to do with their real portfolios.
So BankersLab has been extending itself from the classroom into what it calls a Command Center - an always-on, AI-enhanced decisioning layer built on the same simulation core. The framing is deliberate and, for a fintech in 2026, refreshingly un-hyped: AI here is decision support for humans, not a replacement for them. The pitch to a Chief Risk Officer is that instead of a quarterly retrospective on what the portfolio did, you get continuous, AI-backed recommendations - the future run forward in simulation, surfaced daily, for a person to review and act on. "See the Future. Shape the Outcome" is the tagline, and it is the training thesis grown up: you were practicing on a fake portfolio so that, later, you could run a real one better.
There is a democratization argument threaded through all of this that Katics returns to often. The tools that used to belong only to the biggest banks - machine learning, big data, serious modeling - are now within reach of a small credit union with an internet connection. The bottleneck is no longer access. It is knowing how to use any of it well, which is precisely the gap a simulator is built to close. You can hand a mid-sized lender an AI model, but you cannot hand them the judgment to trust or override it. That, you have to practice.
None of this makes BankersLab a large company. It is a small, distributed team - on the order of a dozen people, spread across California, Singapore, Bangkok and Dubai - that raised a modest $750,000 seed round in 2016 and has otherwise grown the unglamorous way, by being useful to banks in a lot of countries. But the shape of the bet is clean, and it has been consistent for well over a decade: people don't learn finance by being told about it. They learn by running the bank. BankersLab just makes sure it's a bank that can't actually fail.
