Comment by djhworld

8 hours ago

I've been beancount'ing for years now

As we've crossed into the new year I've switched to a similar directory setup as the OP with 1 file per year. Previously I just had one file that was from 2022 which ended up being like 2 million lines of text, which was starting to bog down the emacs plugin.

What I appreciate the most about this approach to personal finances is it just tracks everything. Investments, pensions, RSUs, bank accounts. You could even go as far as accounting for any resource that's modellable, e.g. energy usage in kwh vs. bills. I probably wouldn't go that far though :D

Also you can build a bunch of tooling around it too, with the advent of LLMs my toolset for beancount management has expanded quite significantly. Most recently I got claude to rewrite my transaction rules engine https://djharper.dev/post/2025/08/19/using-llms-to-turn-scri... into something nicer with a UI. This would have taken days to build in the before times, and I probably would not have bothered because it's overkill for 1 user (me)

I have my electricity bills tracking into a KWH commodity. It has been... effectively useless :D.

  • I track my propane in an LPG commodity at a fixed price per season. It saved me about $100 once when a transaction wouldn’t balance. I was accidentally partially charged for a short load delivery on one of my tanks at almost double the rate. Even if it seems silly to track at this fidelity in the moment, I wouldn’t have caught this tracking USD alone. Billing mistakes happen and can be costly!

    • Nice! That sounds really useful; in my case the KWH usage (and price/KWH) I pull directly out of the ConEd bill, so my only chance to notice those sorts of things would be post hoc looking back in time for big jumps in rate or usage I think.

      But good to hear the positive story side for this.

  • I struggle with tracking the actual cost of my energy usage between changing electric rates, the various solar costs, SRECs, different loans and credits and incentives to be able to make an intelligent decision on what the benefit or cost is of cutting an appliance or adding something new. It’s a lot.

    • It is a lot.

      And when it’s fragile even when working. The cost per unit changes with limited notice in various ways (line rate, unit cost, time period that various rates occur, the day, ‘free’ power bonuses etc).

I'm curious about this as I've thought about building personal software to manage my data (such as finances) via LLM, however I'm apprehensive about actually supplying that data to the LLM to help build the tool. Is this what you're doing? Now that I think about it, I could probably 'anonymise' the data with a local model...

  • Nah not exposing the data to the LLM, just building the tools to help me manage the data, usually python scripts

  • Just go for it in your favorite programming language, it's a nice exercise and in the meantime you can learn about (double entry) bookkeeping.

    I did that for myself and it took me at most three months of parttime working on it, resulting in a bespoke solution. No need to keep paying some SaaS or fiddling around in Excel. One still needs to learn the bookkeeping system though, but that's true for all options. A SaaS is not magically make that go away.