panda_abyss

I would never admit this happened to me

I'm sure the number is higher than 2

All I’ve got is a nubby yubikey nano

To be fair, I’m not sure why firebase even has a public access option. That’s a recipe for issues.

Though if it’s anything like Google Cloud Store, they hopefully make it very clear that your bucket is public.

They all got $200M last week

But anything the US feds contracted them for, like building data centres, they have to comply or they face penalties and have to pay all the costs back.

10 days ago, a week before this was announced, they awarded $200M contracts each to Anthropic, OpenAI, Google and xAI

This doesn’t doom the public versions, but they now have a pretty strong incentive to save money and make them comply with the US governments new definition of truth.

What do they do with the daddy computers? All of mine only have female ports.

The first few iterations were good, when they went mass market they let quality go.

Yeah, sadly SpaceX and Tesla are both very promising companies primarily held back by Elon Musk.

Oh wow, I didn’t know you were having these problems with him.

But since you’re in my office let’s start you on a PIP and I have to write you up for this hostility against a fellow employee.

Batch process turning unstructured free form text data into structured outputs.

As a crappy example imagine if you wanted to download metadata about your albums but they’re all labelled “Various Artists”. You can use an LLM call to read the album description and fix the track artists for the tracks, now you can properly organize your collection.

I’m using the same idea, different domain and a complex set of inputs.

It can be much more cost effective than manually spending days tagging data and writing custom importers.

You can definitely go lighter than LLMs. You can use gensim to do category matching, you can use sentence transformers and nearest neighbours (this is basically what Semantle does), but LLM performed the best on more complex document input.

One built a bunch of local search tools with MCP and that’s where I get a lot of my value out of it

RAG workflows are incredibly useful and with modern agents and tool calls work very well.

They kind of went out of style but it’s a perfect use case.

The tool isn’t returning all code, but it is sending code.

I had discussions with my CTO and security team before integrating Claude code.

I have to use Gemini in one specific workflow and Gemini had a lot of landlines for how they use your data. Anthropic was easier to understand.

Anthropic also has some guidance for running Claude Code in a container with firewall and your specified dev tools, it works but that’s not my area of expertise.

The container doesn’t solve all the issues like using remote servers, but it does let you restrict what files and network requests Claude can access (so e.g. Claude can’t read your env vars or ssh key files).

I do try local LLMs but they’re not there yet on my machine for most use cases. Gemma 3n is decent if you need small model performance and tool calls, phi4 works but isn’t thinking (the thinking variants are awful), and I’m exploring dream coder and diffusion models. R1 is still one of the best local models but frequently overthinks, even the new release. Context window is the largest limiting factor I find locally.