[kaydub]

Funny, the element of ego I'm seeing are the devs that can't accept that LLMs write perfectly good code at a faster pace. It's always "slop" this or "vibe coded" that. Meanwhile I've personally shipped multiple apps and professionally we've done HUGE things in these past 6 months using LLM tools.

[tegiddrone]

Sounds like the aforementioned rockstar dev may have found their way to comment and miss the point.

When I was less self aware, I'd say similar things: my coworkers/manager can't accept that I write perfectly good code at a faster pace. Doesn't matter whether it's good or bad code-- the coworkers have to collectively understand it in order for it to be maintainable.

[habinero]

Oh really? Link them.

[kaydub]

Not sure what I'm going to link to in my professional life, it's all private repos. We have a large microservice architecture in AWS. We've made major changes and the LLMs have helped every step of the way. Migrated our whole auth system from being regional, resulting in customers being limited to specific regions or managing their multiple accounts across regions, to a global authentication with data residency built into accounts so that they login to what looks like a global app now. No more uk.example.com, eu.example.com, ca.example.com and us.example.com. It's now just example.com, but data residency (and currently data processing too, but that's another thing we're working on) stays the same. This isn't some small app, this is the main product at a company with nearly a Billion in ARR.

For personal projects, only one thing could I actually show - vaicayo.com

This app was originally started pre-llm. I just wanted a centralized place for my wife and I to organize our vacations. I built out the basic crud for that a long time ago and we used it, but then llms got popular. I first integrated it into the app, now I could get it to generate vacation plans completely with the llm. It did okay.

Somewhere between 3-6 months after building out llm features claude code released or I became aware of it enough to start using it.

I rebuilt the whole thing. So what was once a spring boot app running in AWS ECS with alb + cloudfront (and yeah, some sqs+lambda python processes, s3 buckets for storage, aurora rds for db) is now a super lean apigw + lambda + dynamodb stack with a svelte frontend. I went from paying $20ish a month to $0. I also built a flutter app for android and ios that utilizes the same backend as the frontend webapp with feature parity. It's not really a production app, but it could handle it fine, it's completely cloud native and hands off to operate. And let me be clear about the rebuild, it's not a 1 for 1, I think all that I kept was the domain model.

For features, it's mostly just a crud app. You plan your vacation in it. Outside of that there's the ability to take pictures with the phone app that auto-sync (there's a bit of complexity here, the phone app only syncs pics when it's on wifi unless you change settings to allow syncing on cell network, images are cached locally with a ttl, uploads will run in the background and queue up, uploaded images automatically resize/compress and generate thumbnails from an s3 event lambda). The other more complex item is email ingestion, when I get a reservation email I forward it to the app and it automatically processes it (the only place I use an LLM in the app right now). If it can't discern what the email is or which vacation it goes to it goes in a travel inbox so the user can route it to the right vacation and object type. Oh I guess the paywall is a little complex. I don't advertise this thing but I have left user registration open if anybody ever wanted to use it. I did lock the email parsing behind a paywall and I limit photo uploads for free accounts.

So frontend, backend apis, flutter app for android and ios all built 100% by llm.

For code quality, the frontend is okay, the backend is pretty clean but could use some things split out into domains, and the flutter app has some work to do (recently found a couple code smells). But there's no issues adding features, shipping, etc. I literally finished adding some of the photo syncing features in the flutter app last night.

Other than that, I have created 5+ other personal projects that I wouldn't have without an LLM. I've made 2 godot games, a private scribbl.io clone with a few extra features that myself and my coworrkers play as a "happy hour" on fridays, tons of ci/cd build stuff and terraform (using my own personal modules, so I guess that's not 100% llm coded if you consider the modules), and then a couple other simple apps.

[munksbeer]

I enjoyed hearing about the stuff you've done.

Would you mind talking about your experience making the 2 godot games? Which agent and model did you use? I've been wanting to dabble here and interested to hear stories.

[kaydub]

Got distracted and didn't touch on the godot specifics.

I start the project in godot, then I talk to an agent to define my game and give details. I tell it to have a conversation with the end goal to be building a CLAUDE/AGENTS/GEMINI markdown file for future agents to help build the game.

I made a [single player top down naval game](https://www.youtube.com/watch?v=FQUTbeQ8-oA) and an online multiplayer game (that I'd like to keep to myself for now).

I mostly used free assets for the artwork. I made some myself in aseprite. I utilized LLMs a tiny amount to generate some assets, but typically those are more of a jumping off point than game ready art assets. This has been what's slowed me down the most in making games.

[kaydub]

I use any of the 3 big frontier models; claude, chatgpt, gemini

I only use the cli tools; claude code, codex, gemini cli (which is being phased out for antigravity cli soon)

I've generally had the best luck and enjoy working with claude the most, though that may be because that's the model we use at work (from aws bedrock, we pay per token and always have). Gemini and codex have flip flopped on 2nd place, though right now I'd lean Codex.

On my personal stuff I typically use codex or gemini because those two I have subscriptions, claude I still pay per token but through anthropic API like 90% of the time and 10% of the time through bedrock.