Anthony Figueroa

CTO and Co-Founder at Rootstrap. Engineer. Blockchain and AI Enthusiast. Born in Uruguay 🇺🇾.

[Anthony Figueroa @anthfb]

@BriiHD Depende la versión necesitan siesta también

[Anthony Figueroa @anthfb]

@nachomdeo me preocuparía las herramientas ilegales

[Anthony Figueroa @anthfb]

@marcel_mordezki @AldoLema_uy Tal cual. Muy inocente pensar que “seminarios de ética” van a cambiar comportamientos.

Nunca es un problema de ética, siempre de incentivos.

[Anthony Figueroa @anthfb]

@AgustnHenderso1 Es lobby de manual esa decisión

[Anthony Figueroa @anthfb]

@sama That one is a non-profit, for now

[Anthony Figueroa @anthfb]

@Shem_Infinite Obviously a magic carpet

[Anthony Figueroa @anthfb]

@HormuzLetter It’s the Strait of Schrödinger, open and closed at the same time.

[Anthony Figueroa @anthfb]

@danitoszwarc @ElBioLoco @miriamgonp @javilop Hay modelos específicos entrenados en cosas médicas, como MedGemma de Google. Seguramente tenga un rol que cumplir en ese proceso. 🤔

[Anthony Figueroa @anthfb]

@_bruno_t Por más que se intente, mucha gente no logra pensar en términos de oferta y demanda porque le resulta más intuitivo centrarse en intenciones, y le cuesta ver cómo las decisiones de muchas personas interactúan y terminan formando los precios.

[Anthony Figueroa @anthfb]

@DotCSV Tendemos a proyectar el futuro con los parámetros y limitaciones del presente, porque nos cuesta imaginar al mismo tiempo una mejora tecnológica continua, y potencialmente acelerada, sesgando nuestro análisis sobre cómo evolucionarán las cosas.

[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.

[Anthony Figueroa @anthfb]

@_chenglou Amazing work

[Anthony Figueroa @anthfb]

@dosinaga2 El que sabe usar polymarket más allá de mirar predicciones, sabe usar una VPN

[Anthony Figueroa @anthfb]

@ealmeida Lamentable la deriva que tomaron, pero supongo que es comportamiento humano básico nomás.

[Anthony Figueroa @anthfb]

@franperez_co La parte interesante es que quizás en realidad todos los patos eran tuyos y tus 250 seguidores te los estaban guardando. Y de paso podes decir que esos 250USD son pérdidas de capital mientras te guardas los patos y un buen porcentaje de los pagos.

[Anthony Figueroa @anthfb]

@nachomdeo Estás en tu mejor momento 🤣

[Anthony Figueroa @anthfb]

@levelsio +1 on your request. iFood almost seems designed to make things impossible for foreigners.

[Anthony Figueroa @anthfb]

@matteopelleg Collateral as a service

[Anthony Figueroa @anthfb]

@levelsio @marcospereeira @grok In Spain it worked WAY better than any optic fiber provider.

[Anthony Figueroa @anthfb]

@QueOpinaUY @PhDenLogica No quieren mantener la tenencia. Bajaron la tasa de política monetaria de 7.5% a 6.5%. Supongo que quieren que se vaya más hacia algo cercano al 5%.