The Church of the TigerBeetle

The Church of the TigerBeetle

Tech Evangelism

Posted on August 3, 2026

by Adam Wespiser

The seats of the Institute of Contemporary Art in Boston are somewhat comfortable.<br>A plush base you can almost sink into, but only enough elbow room for you or your neighbor, not both.<br>By the end of an hour, you’ve explored the full space of position adjustments, and have the type of “settle in” feeling I only feel on flights.

The final talk of Systems Distributed 2026, titled Mission-Critical, by Joran Dirk Greef, opened with a definition so efficient it nearly circled back on itself: mission-critical software, we were told, is software that performs a critical mission.<br>Got it. That was quick, here we go dinner!<br>However, the talk continued, and came to TigerStyle, impressive logo and all, with the claim, delivered with absolute clarity, that this engineering philosophy was what made TigerBeetle possible.

Next came several claims about TigerBeetle, among them that a large share of the transactions on a national exchange run on it, capped with a mission statement about powering the world’s transactions.<br>The exchange number is real and genuinely impressive; it was the framing around it, the grand mission delivered as an applause line, that gave me pause.<br>Before I could settle with the idea, we were looking at the Jepsen report, a testimonial of sorts, highlighting the lines that affirm TigerBeetle made the right calls.

It was getting to be a lot, especially with the thank-you interlude and the applause for contributors with enough social pressure, or genuine admiration, that everyone around me clapped.

Then the music started, playing over a TigerBeetle sponsored race car, and my only thought was that this was a remarkable way to drive all day and end up in the same place.<br>If I’m not getting something out of a talk, I don’t owe anyone the social obligation of staying (and neither do you).<br>I excused myself from the middle of a row, and spent the rest of the talk in the lobby staring at an installation in amazement.

What I witnessed wasn’t a normal tech talk, it was a sermon from the Church of the TigerBeetle, complete with origin myth, testimonials, and orchestral music, all leaving you with the impression that their way was indeed the best.

It’s tech evangelism, and that scares the shit out of me.

The problem isn’t that TigerBeetle is making bad decisions with their tech, their deterministic testing, viewstamp replication protocol, and use of money and influence for good causes is impressive.<br>The problem is the claims made on faith.<br>Claims made without introspection, without proof, presented to you one after another without a moment in between to digest and integrate them into your current belief system, and set to emotional background music.<br>Are we gearing up to storm the Bastille here, or are we trying to build software that works for our end users?

The question shouldn’t be “Is this organization inspiring?”, but “Can we steelman these claims?”<br>Without that, we’re walking around with ideas, but not the rigor we need to defend them.<br>Not to ourselves, not to our teammates, and not to our organizations.<br>That’s a problem.<br>I’m not trying to dunk on TigerBeetle.<br>What worries me is the mode of persuasion: the presentation asked the audience to admire and believe before it equipped us to evaluate.

If I pay $500 (or my company paid that to send me), what good can I do if I believe the hype but don’t carry the weight of the evidence?<br>I want to believe that TigerBeetle is great, but faith without receipts is useless in most organizational cultures.

The other place I’ve seen tech evangelism is in Haskell, specifically the belief that strongly typed programs are a better way to build software.<br>It’s a claim that tends to be held in proportion to your knowledge about types, but it’s lacking evidence that it works.<br>In each of the three companies I worked for, Haskell started as the solution to software, but was eventually blamed for the company’s problems.<br>The blame was not fully deserved, but the fervent belief in the language made discussions of language tradeoffs difficult when being a Haskeller was part of your identity.<br>That’s not to say you shouldn’t get to pick your tools, or choose where you work based off those tools, just that a purely affinity-based identity is not a solution to technical problems, and it often gets in the way.

Evangelism is considered harmful as an organizational technology.<br>Even if everyone in that room (except me) is already armed with the evidence, rallying the faithful trains people to hold beliefs by faith, not by reasoning.<br>An org that rallies by faith teaches people to default to faith — so when a genuinely hard question comes up, like “should we enforce a coding standard?”, they reach for conviction instead of evidence: it worked for TigerBeetle, they do X.

Second, evangelism suppresses dissent — and my revulsion for the style is a feature of it, not a...