Hacker News: adamkurkiewicz

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 20:51:39 +0000

Sure, I'm at adam@vineyard-finance.com

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 19:54:10 +0000

I don't know how much a tax lawyer/ account would quote for this in California, but I saved 3000 EUR/ tax year on a German Steuerberater in a similar way.

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 19:48:48 +0000

Hey, the author of the benchmark here.

The benchmark data was prepared in April 2026 (when I was manually doing our VAT return with my co-founder). The invoices were indeed found manually.

Currently we're using a custom "invoice searcher" built on Kimi 2.6 (in our testing several weeks ago it outperformed Opus 4.7; it was just more persistent).

Ultimately, I still verify everything manually after the model is finished fetching invoices for the month -- but it's a great help to have all the invoices already found (usually correctly).

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 19:40:23 +0000

I've submitted my German taxes this year using a mix of Claude 4.6 and Claude 4.7, with lots of manual checking. The German Finanzamt granted most of the things I listed in the tax return (they send you an official letter by post) -- I did have to appeal for one of the items though (again using Claude, this time 4.8 ).

The most important thing I've found is to ask Claude to thoroughly audit the reply (to find all hallucinations). I usually ask it to give me an enumerated list of all facts and all legal cases quoted, and then I give it to a new instance to carefully validate each one.

Newer models are getting much better at not hallucinating German case law though :)

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 19:29:46 +0000

We've got integrations with major UK banks. Curious if you'd like to use a polished product or be more interested in bank-feed-as-an-API type of use case? What banks do you use?

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 19:24:28 +0000

Info on the founders coming soon -- we're just going public with this.

For slightly out of date founder bios (both Adam and Iva) were also co-founders here:

https://www.biomage.net/our-team

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 19:21:58 +0000

Very cool!

We've used the following CLI to do the freeagent upload:

https://github.com/anjor/freeagent-cli

How are you dealing with finding the receipts? Would you like to try a receipt finder that grabs them from your mailbox/ google drive?

New comment by adamkurkiewicz in "GLM 5.2 is nearly as accurate as a human book keeper"

adamkurkiewicz — Thu, 09 Jul 2026 19:16:32 +0000

Hey, author of the blog post here.

I was one of the human book-keepers for this benchmark (the preparer; my co-founder verified the VAT submission once ready), and given that at the time of doing this I knew I was eventually going to use this data for evaluating the models, I was super careful. So I guess this is a "good book-keeper". In the previous company our book-keepers made lots of mistakes; some serious enough that we had to restate our company's accounts.

GLM 5.2 is nearly as accurate as a human book keeper

adamkurkiewicz — Thu, 09 Jul 2026 18:29:41 +0000

Article URL: https://toot-books.com/blog/glm-5-2-vat-benchmark

Comments URL: https://news.ycombinator.com/item?id=48850414

Points: 222

# Comments: 119

New comment by adamkurkiewicz in "Send (real) money with Python in 49 currencies"

adamkurkiewicz — Wed, 24 Sep 2025 13:34:56 +0000

This guide shows how to send real money (such as USD, EUR, GBP, and 46 other currencies) using the Python programming language. Money is sent using bank transfers between bank accounts, and currencies are converted at competitive rates for international transfers (so one can hold money in e.g. EUR and pay it out in e.g. USD).