👨🏻‍💻 Luis Assardo

23.2K posts

👨🏻‍💻 Luis Assardo banner
👨🏻‍💻 Luis Assardo

👨🏻‍💻 Luis Assardo

@luisassardo

Exploring digital threats, hostile environments, how to deal with them and their impact on our society. | retired firefighter | rōnin | 🇬🇹🇩🇪

Berlin, Germany Katılım Kasım 2008
918 Takip Edilen8.6K Takipçiler
Yaw Nyame, MD | Mind & Body
If you live in Germany as an immigrant tell me something living in Germany has thought you ...?
English
67
22
123
149.2K
👨🏻‍💻 Luis Assardo retweetledi
Storch_i
Storch_i@Storch_i·
Seltsam: heute gibt es gar keine Sondersendungen bei NIUS, BILD & WELT, auch keine Titelgeschichten, keine Leitartikel, keine Interviews mit Beteiligten o.ä. zu dem Machten-Täter von Görlitz. Das hat exakt einen Grund: der Täter ist ein deutscher Nazi.
Deutsch
167
723
3.9K
133.1K
👨🏻‍💻 Luis Assardo retweetledi
World's Amazing Things
The beauty of a Quetzal
World's Amazing Things tweet media
English
28
642
4.4K
53.4K
Awasss
Awasss@awitaenelzaguan·
Qué miedo que la decisión de mudarme a Alemania, tan aleatoria y tomada a la ligera, me haya marcado así de profundo y que, aún peor, no pueda revertirla o al menos regresar las cosas a un estado similar al que estaban antes de venir.
Español
67
49
844
106.1K
👨🏻‍💻 Luis Assardo
I started to build something like that with thousands of work related documents downloaded over the years. Keep using the “folder structure index” is crazy. Mine gives me a simple dashboard where I can search manually, via bot and soon offers relevant data while working a consultancy. Soon will be at github.
👨🏻‍💻 Luis Assardo tweet media
English
1
0
3
1.4K
Luis Garicano 🇪🇺🇺🇦
Two things: 1. If you care about AI and you don't follow Karpathy, you are making a major mistake. He is a huge provider of public goods. 2. This idea is genius, specially for all academics. I have not implemented it. But it is my next project.
Andrej Karpathy@karpathy

LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So: Data ingest: I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them. IDE: I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides). Q&A: Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale. Output: Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base. Linting: I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into. Extra tools: I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries. Further explorations: As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows. TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

English
57
154
4.9K
2.7M
👨🏻‍💻 Luis Assardo retweetledi
NASA Earth
NASA Earth@NASAEarth·
That's us! 🌍 The Artemis II crew captured beautiful, high-resolution images of our home planet during their journey to the Moon. As @Astro_Christina put it: "You guys look great."
NASA Earth tweet media
English
3K
42.8K
218K
8.5M
👨🏻‍💻 Luis Assardo
👨🏻‍💻 Luis Assardo@luisassardo·
@handsomeADEDD @YFIHQ without fire pressurized gas spreads really fast and you will have an explosion. Thats why the protocol is to close the valve, os seal the leak. Meanwhile someone else cool down the gas container.
English
1
0
2
241
YFI 𝕏
YFI 𝕏@YFIHQ·
Heroes.
English
136
359
10.7K
2.7M
👨🏻‍💻 Luis Assardo retweetledi
Beauty of the Catholic Faith
Beauty of the Catholic Faith@advoluntas·
♰🇬🇹 Palm Sunday in Antigua, Guatemala 🇬🇹♰ 🎥 Solemn Procession at the La Merced Church on 29 March 2026 in Antigua, Guatemala 🇬🇹
English
9
322
1.5K
264.6K
👨🏻‍💻 Luis Assardo
👨🏻‍💻 Luis Assardo@luisassardo·
Kerry ya es campeón en comida y chucherías chapinas. Desde botonetas hasta guayabitas 😍. Me va emocionar verlo comer de todo en Guatemala. Un tipazo, chapin de corazón.
Español
12
38
345
11.1K
👨🏻‍💻 Luis Assardo retweetledi
SEGA
SEGA@SEGA·
We’re excited to announce that we’re making our SEGA logo 3% more blue. We hope you enjoy this upgrade. Please take a moment to familiarise yourself with the new look.
SEGA tweet mediaSEGA tweet media
English
3.1K
7.9K
88.2K
2.8M
👨🏻‍💻 Luis Assardo retweetledi
Denny’s
Denny’s@DennysDiner·
Denny’s tweet media
ZXX
274
1.6K
15.4K
250.8K
👨🏻‍💻 Luis Assardo retweetledi
DoorDash
DoorDash@DoorDash·
for legal reasons this is a joke
DoorDash tweet media
English
227
1.2K
16.3K
969.2K
👨🏻‍💻 Luis Assardo retweetledi
McAfee
McAfee@McAfee·
McAfee tweet media
ZXX
40
384
3.3K
67.7K
👨🏻‍💻 Luis Assardo retweetledi
Montréal
Montréal@Montreal·
Montréal tweet media
ZXX
127
1K
11.6K
204.2K
👨🏻‍💻 Luis Assardo retweetledi
Domino's Pizza UK
Domino's Pizza UK@Dominos_UK·
Domino's Pizza UK tweet media
ZXX
2.7K
16.4K
226.5K
10.6M
👨🏻‍💻 Luis Assardo
👨🏻‍💻 Luis Assardo@luisassardo·
@ciudadanoalt Propones que siga sufriendo en nombre que todo debería funcionar como debe y porque piensas que vivir es una obligación. Ojalá y todo fuera tan fácil.
Español
2
0
7
719
Javi Cabello
Javi Cabello@ciudadanoalt·
¿Lo mejor que pueden hacer el Estado y una sociedad decente por una joven deprimida, sin ganas de vivir y sin ninguna enfermedad terminal… es matarla? ¿Qué precedente estamos generando?
Español
661
2K
7.1K
105.6K