Vicki Boykis

Related: Vicki Boykis has a great blog[1], particularly for those trying to get their feet wet in the ML engineering space—I really found "How I Learn Machine Learning"[2] super useful as a template to tackle new ML problems, as someone who is new to the field. [1] https://vickiboykis.com/ [2] https://vickiboykis.com/2022/11/10/how-i-learn-machine-learning/

That is essentially correct. You take an object and "embed" it in a high-dimensional vector space to represent it. For a deep dive, I highly recommend Vicki Boykis's free materials: https://vickiboykis.com/what_are_embeddings/

Since publishing this I've found a few additional resources that are really useful for understanding embeddings at a lower level (my article is deliberately very high level and focuses mainly on their applications). Cohere's Text Embeddings Visually Explained: https://txt.cohere.com/text-embeddings/ The Tensorflow Embedding Projector tool: https://projector.tensorflow.org/ What are embeddings? by Vicki Boykis is worth checking out as well: https://vickiboykis.com/what_are_embeddings/ Actually I'll add those as "further reading" at the bottom of the page.

I remember this post[0], here, a while back. I’ve been shipping (as opposed to coding), for my entire adult life, and I’ve learned (the hard way, of course) that “ship” is always at least a couple of clicks back from “bleeding edge.” I’m in the home stretch of an app that I’ve been developing for the last year and a half, or so. The backend is written in PHP, and the frontend is a “classic” UIKit/Storyboard/MVC app, as opposed to a SwiftUI/Combine/MVVM project. It’s fairly ambitious, and, when I started, I was not confident that “the bleeding edge” would work (it might have, but I didn’t know that it would). There was no question that the classic patterns would work, so I picked them. It’s coming along great; far better than I had originally envisioned the project. It is ultra-high-quality, fully native, with only one small external (meaning that I didn't write it, myself) dependency (the backend has zero dependencies), has many capabilities that have only appeared in the last couple of OS releases, is easily localized, conforms to multiple device configurations, dark mode, accessibility features, etc. I read (here), about a well-known app that had been a highly successful classic native app (probably written in ObjC) that was supposed to be rewritten in SwiftUI, but the project failed, and was eventually shipped in Electron. That’s kind of my worst nightmare. I suspect the developers had to put clothespins on their noses, for much of the project. I will be releasing…

don't think she has posted here, but Vicki Boykis blogged about this today: https://vickiboykis.com/2026/06/15/running-local-models-is-good-now/

Vicki Boykis wrote a small book about it: https://vickiboykis.com/what_are_embeddings/

Article was deleted and reposted as '20 Years Of YC' https://vickiboykis.com/2025/03/17/20-years-of-yc/

Off the top of my head - try to implement techniques that are doable on home hardware like the one described in OP (requires some $$$ investment) and give feedback or contribute to documentation / guides - learn about different techniques and do educational writeups or documentation (like https://vickiboykis.com/2024/02/28/gguf-the-long-way-around/) - build a tool / library that wrap academic techinques and expose them more easily to end users (like A1111 or comfyUI for stable diffusion) Anything that can translate the high end research down to something a moderately technical user can use or learn from is a gigantic win

IT Runs on Java 8 https://vickiboykis.com/2019/05/10/it-runs-on-java-8/ hn discussion https://news.ycombinator.com/item?id=19877916

The pull quote in the center is so divine, tells the entire story of where we were vs where we are: > Instead of working on the core of the code and focusing on the performance of a self-contained application, developers are now forced to act as some kind of monstrous manual management layer between hundreds of various APIs, puzzling together whether Flark 2.3.5 on a T5.enormous instance will work with a Kappa function that sends data from ElephantStore in the us-polar-north-1 region to the APIFunctionFactoryTerminal in us-polar-south-2. From Vickyli Boyis a couple days ago, https://vickiboykis.com/2022/12/05/the-cloudy-layers-of-modern-day-pro…

You might want to check out Automattic (the makers of Wordpress) and their P2 tool - https://vickiboykis.com/2021/07/17/writing-for-distributed-teams/