Ville Säävuori

Most of my Django projects run on a dedicated VPS using Docker compose behind Traefik. Simple and fast enough for most situations.


A larger project grew to ~50s downtime between deploys. Claude + codex wrote me a simple blue/green deplo; near-zero downtime in ~200 LOC 🎉

Thinking of times ~15 years back when I worked in an agency. It used to take DAYS to produce a technical plan even for a simple MVP. You still need to understand the clients needs etc but I bet that process alone can be 90% automated. Hell, just sell MVP as a (self-)Service 😃

I had a very well working workflow for this using the plan mode (cc even has a keyboard shortcut opening the plan in your editor!), now the process is much clumsier as I have to guide Opus to use the ask questions tool AND I have to manually manage the plan files. Such a dumb step backwards!

I'm realizing that it's not Opus the model but Claude Code the vibecoded product that puts me off of using it anymore. Hard to understand how it got so much worse in such a short amount of time. I think its because I relied on plan mode. Oh well.

I resisted *checks notes* 25 hours before setting up Django for my clanker.

I'm immediately worried that I might regret this.


..only because I can see this swallowing *so* much time! (And no, its not That clanker, this one is all homegrown ❤️)

The frontend config hell is such a PITA. Some random package upgrades made type checking take over 90s from ~25s in a monorepo vue/TS project with 3 packages.


Well, after a metric shit-ton of config tweaks and added TS config json, Claude dropped it to 3s 🫠

Not sure whats happened but for the whole day today Claude has been just thinking, spending tokens and writing absolutely nothing or maybe one line after ~15 minutes just eating tokens. Thanks, Claude -- not _quite_ the way I'd like to spend my money 🫠

I can't get used to how often gpt-5.2-codex fails. It's excells at many tasks but also fails spectacularly at times.


Just now it wrote five (!!) identical ~10 line helper functions across ~10 test files instead of writing it once and importing 🤯

Has anyone implemented a simple "this is what we've been working on recently" file list for agents? Just maintaining a list of last ~25 paths touched would speed up starting sessions a lot.


Been too lazy to try to build this but it would only need a small shell script 🤔

The Sorcerer's Apprentice comes to mind so many times when I ask Claude to do something and before I remember to check on it, it has completed something WAY more complex that I intended.


Successfully. All by itself. And it's done. And I just went to pee for a second 😅

Ahh. The oldest fail in the book; managed to fill prod server disk -> dbs failed 🤦‍♂️


I even had a check for this but it failed too bc it wasn't properly tested! 🤦‍♂️


..oh well, Claude wrote me better checks plus a nice dashboard view for admin so hopefully not happening again 😅🎉

No matter what I do gpt-models keep lying to me today.


"- Here's the answer: ...
- Are you ABSOLUTELY sure this is 100% correct?
- [fucks around for ages] .. Yes.
- Prove to me that this is correct.
- [fucks around for ages] .. I was wrong, here's the correct answer:"

Webdev tooling is fantastic nowadays. It took me less than 15 minutes to write a pretty complex e2e test using Playwright that tests (in a single self-contained test) user registration, login, and some non-trivial realtime two-user chat functionality 🤯

`ArtistProfilePhotoUploadResponseSchema` in my Python code -- really, gpt-5.2-codex?


What's the coding equivalent of forcing someone to smoke a full pack of cigarettes after getting caught smoking? I want to force my codex to read Java code as punishment.

Asked Claude to explain what all of the GitHub actions did that @simonw published the latest versions for:


https://claude.ai/share/15ce3469-07e7-4d95-b5f9-a6eae524d6fa


TIL there's for example a "first-interaction" action that allows you to greet people when they first interact with your repo!

Claude thinking for a few seconds and then claiming "I now have a comprehensive understanding of the codebase" always feels a bit defeating.


We're fast approaching the day when that statement will be true 😬

Given how absolutely horrible the code that ALL current LLM models write is from a security perspective, I'm still amazed we haven't seen more major incidents yet.

The new gpt-5.2 xhigh seems very good at planning and writing specs.


Unfortunately codex is totally useless for any coding bc for some reason it just does not work at all with uv (which seems weird as all other tools work fine) 🤷‍♂️

Feeling like an old man yelling at clouds bc I get frustrated about people who talk about blogs when meaning blog posts.


"I wrote a newspaper about the spoiled youth" 🤔

I love it when projects host their own "awesome"-lists and merge in repos with 1 commit and no activity in two years -- pretty "awesome" 😂

"Create TypeScript interface for the shape of this JSON" is one of my favorite prompts.


It's *so* useful and saves so much time and head aches when dealing with random third-party APIs 🎉

Can't for the life of me understand how people get any value out of gpt5-codex. I'm really trying to give it a chance but it's just total garbage all around. It writes *really* shitty code AND is slow as hell 🤔

It's crazy how much automated phishing traffic there is nowadays. Seconds after pointing Tailscale funnel to my Django devserver the logs are full of malicious requests to env files, conf files, known Jira and php paths etc 🥴 (bc it has a known ssl cert)

Ah, yet again the joys of working w the JS hippies;


bump an official Vue ecosystem package from MAJOR.MINOR.PATCH to PATCH+1 and your build just breaks with gazillion of config-related TS issues 😅

a pixelated image of a man making a face ALT: a pixelated image of a man making a face

Developers: a better coding model, please?
OpenAI: pro-users can set special color themes for their chats!

I'd love it to remember better that 99,999% of the time I *only* ever translate between two languages. (Most of the time they can be read from my browser but also from my Kagi profile.) Having to manually select those every single time is annoying AF but the service itself works great 🙏

The thing is, I'd *never* spend the day or two needed to figure out the root cause (sqlite not always closing properly between the async tests), and having totally unrelated warnings poison the test output just seems worse than having swept it under the carpet w magic 🤷‍♂️

Not sure what I think of the fact that I let Claude to clean up warnings from a pretty complex pytest suite that runs async Django WS tests w/ Sqlite. I *know* the tests are working and no warnings now .. but conftest.py has 100+ lines of horrible magic 🤔

I recently used ULIDs with a Django backend, like them a lot; much more compact than UUiDs, orderable, include a timestamp and easy to serialize 👌