While GitHub has been busy losing its last nine of availablility, I’ve been thinking about how the
internet used to be.
Not the internet people talk about from the 90s, but the internet that we used to have even 10-15 years ago. This was the heyday of...
I stumbled across this report from NASA, “Elements of Engineering Excellence”,
published in 2012,
The inspiration for this paper originated in discussions with the director of MSFC
Engineering in 2006 who asked that we investigate the question: “How...
If you love building things, and the process of building is just as important to you as the result itself, it’s not unreasonable that you’re in a slump these days.
The world is telling you that your thinking process is extraneous, unnecessary, and...
Recently, I got nerd-sniped by this exchange between Jeff Dean and someone trying to query 3 billion vectors.
I was curious to see if I could implement the optimal map-reduce solution he alludes to in his reply.
A vector is a list/array of floating...
Jeune fille lisant une lettre à la bougie, Jean-Baptiste Santerre, 1700
Machine learning engineers spend their lives alternating between two states: staring at tqdm progress bars during model training and staring at error logs during model...
I respect quality software and the people who write it. And, I’ve invested years of my life in working on becoming one of these people (even if the journey has been long and hard and has involved lots of YAML). I have seen and used code written by...
There is a very vigorous debate happening online right now around what shape evaluation for LLM-based products should take. I don’t want to rehash all of it, other than saying that if you are building any applications with with non-deterministic...
A few years ago, I wrote a paper on embeddings. At the time, I wrote that 200-300 dimension embeddings were fairly common in industry, and that adding more dimensions during training would create diminishing returns for the effectiveness of your...
There comes a time in every woman’s life when she only wants one thing: for her mininmal static site to finally have some of the same features that dynamic blogging platforms do, namely search.
So now I’ve implemented search on this blog, you should...
One of my favorite AI dev products today is Full Line Code Completion in PyCharm (bundled with the IDE since late 2023). It’s extremely well-thought out, unintrusive, and makes me a more effective developer. Most importantly, it still keeps me mostly...
I saw recently that YCombinator celebrated its 20th anniversary.
Hacker News is slightly younger, but to me the two go hand in hand.
As far as I can tell, I actively started reading Hacker News around 2011. I don’t remember how I heard about it. It...
Icon by iconixar
Since I signed up for Bluesky last year,
I’ve been wanting to make something using the AT Protocol that the platform is built on top of.
I finally had a chance to do it over the holiday break and built GitFeed, a small Go app that...
If you’re building a new Python web app these days, there’s a good chance you’re using FastAPI. There are a lot of features that make FastAPI easy to get started with. There are also a lot of nuances that take a while to understand. One feature I’ve...
I want to get back into writing more regularly this year, so in light of that, here’s my last year in review.
Evaluating LLMs
Like many of us in tech, I spent a large portion of 2024 thinking about and working with LLMs, but I was lucky enough to do...
There is a lot of debate in the software community around whether LLMs can replace developers. Part of the reason is the way we formulate the problem of what it means to write software. In industry, we still give outsize cultural deference to software...
We keep trying to get LLMs to do math. We want them to count the number of “rs” in strawberry, to perform algebraic reasoning, do multiplication, and to solve math theorems.
A recent experiment particularly piqued my interest. Researchers used...
In the 1800s, before serfdom was abolished in the Russian empire, landowners paid taxes based on how many serfs they had. A census was conducted every few years by government employees traveling across the empire and doing counts; a manual map-reduce...
This is a near-transcript of the talk I gave at PyCon Italia 2024 in May in Florence.
Introduction
Buongiorno PyconIt, grazie per avermi invitata a parlare! Avrei voluta fare tutto il discorso in italiano, ma lo sto ancora imparando.
Per adesso...
Jakob’s Law of UX goes something like this. I, as a user online, spend my time on many sites. As such, when I come to your site, I am already used to the way the other sites work, and I don’t want to learn new paradigms. Some also call these...
We are now in a very weird liminal space in information retrieval for consumers, particularly those attuned to trends in search and working on the bleeding edge of LLMs.
On the one hand, we have the fall of old companies. Broadcast-based centralized...