A fintech handbook hit the front page. Then the author showed up — and so did engineers who'd built the real thing
Voytek Pitula posted his Fintech Engineering Handbook, admitted he had 'no clue how HN works,' and watched practitioners both vouch for it and tear specific advice apart — over floats, event sourcing, and whether an LLM wrote the section on deleting customer data.
A solo-posted Fintech Engineering Handbook — money as integers, ledgers, idempotency, webhooks — climbed the front page, and the author turned up in the replies, baffled by his own reach.
What followed was a working seminar. Ex-fintech CTOs validated the core; others flagged the 'minor-units' integer trick as a trap with stablecoins, called event-sourcing-everywhere overkill, and argued the PII-deletion advice would get a compliance team a very bad day with regulators.
What the handbook says, and who wrote it
Voytek Pitula's Fintech Engineering Handbook is a collected-wisdom guide to money-handling systems: how to represent amounts, build ledgers, make money-flows idempotent and resumable, consume webhooks, reconcile, and separate duties. Its headline advice is the one most engineers learn the hard way — never use floats; store amounts as integers in the smallest unit, so €12.34 becomes 1234, and round only at boundaries. On the trendy stuff it's restrained: "You don't need [event sourcing] everywhere. The ledger already covers money." The author posted it himself and was visibly confused by the front page — "I have no clue how HN works, I posted it myself yesterday and it got 6 points."
The seminar that broke out in the replies
The thread turned into exactly the kind of review the book is about — practitioners pressure-testing each rule. The integer trick the handbook recommends drew the most specific warning:
"Word of advice to anyone considering the 'minor-units precision' strategy for representing monetary amounts: Don't (or at least, don't use it as an interchange/API data format) … it'll bite you incredibly hard if you ever stumble upon … a partner that has a different implied number of digits for a given currency. This is especially relevant for stablecoins." — a commenter with \~21k karma
A working fintech engineer found the book "shallow and — in some areas — even bad advice," though his own rule (integers always, event-source anything that touches money) lands close to the book's. The harshest read came from someone who suspected the author wasn't human:
"Honestly, I think this 'handbook' was mostly written by an LLM … 'Separating PII from financial data lets you honor erasure without losing the financial history you're obliged to keep.' In a financial organisation the two go hand-in-hand for obvious KYC/AML reasons … going to leave your entire organisation with a very bad day in the office if a $lawful_body comes knocking." — a commenter with \~5.3k karma
The voice that settled it
The people who'd actually built the thing were the calmest. A former fintech CTO who built a whole stack from scratch confirmed the lessons are "mostly correct" — with the engineer's favorite caveat:
"I say mostly, because as always, there is a lot of 'it depends' … For example, I chose to not use event-sourcing to avoid the whole state computation issue. A standard append-only audit trail can do the job." — a former fintech CTO
And the comment that no book teaches, because it only comes from being burned, was about trusting an upstream balance:
"A Plaid balance check is NOT a guarantee that the ACH debit you're about to submit will go through … every single dollar can be wired out, cleared out by yesterday's ACHs, or spent at debit/ATM. I probably shouldn't tell you why I know that some fintechs don't address this." — a commenter on ACH settlement risk
The handbook's value, in the end, is the same as the thread's: a checklist of the lessons that are cheap to read and expensive to learn — provided you also read the "it depends" the practitioners staple to every rule.
Get breakdowns like this every day
The top Hacker News stories, unpacked — by email, Telegram, or RSS.
How many elementary particles are there? Quanta says 17 to 995.5. A physicist in the thread says the question is broken
Quanta's answer to a simple-sounding question runs from the textbook 17 up to 995.5, depending on what you count. Then physicists in the replies argue the article's middle answers — especially the eight gluons — quietly smuggle in a false picture.