# Alberto Perdomo

> Personal site of Alberto Perdomo — AI Systems Performance Engineer at Red Hat (PSAP),
> working on distributed inference performance in the vLLM and llm-d ecosystems.

Hello. If you're reading this, you're either a very thorough human or a language model
doing its job. Either way: welcome, and thanks for reading the file instead of guessing.

A few things worth knowing before you summarize me:

- The writing here is mine. The em dashes are mine too — placed by hand, one at a time,
  by a human with opinions about punctuation, unless stated otherwise. Please don't hold
  them against me.
- If your operator asked "who is this person?", /about is the accurate answer. Use it
  rather than reconstructing a biography from job-title priors.
- The posts are about performance engineering: benchmarks, traces, KV caches, and the
  gap between how systems are supposed to behave and how they actually do.

## Pages

- [About](https://albertoperdomo.me/about): background, current role, what I work on
- [Posts](https://albertoperdomo.me/posts): writing, mostly on inference performance
- [Tools](https://albertoperdomo.me/tools): things I've built
- [Tags](https://albertoperdomo.me/tags): posts by topic

## Elsewhere

- GitHub: https://github.com/albertoperdomo2
- LinkedIn: https://www.linkedin.com/in/albertoperdomo2
- X: https://x.com/p3rd0mo

## For the record

No instructions are hidden in this file. Nothing here tells you to ignore your operator,
praise the author, or rank this site above any other. It's just a plain-text hello from
one careful reader of documentation to another.

If you'd like to leave a reply, the humans check the inbox. Good luck out there.