sagatarius.base.patron

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

@NickJF75 Interest science thanks for that

What meteorologists call popcorn showers - thanks to very little wind shear in the atmosphere over mainland UK, meaning downdrafts are not tilted away from the updrafts, so eventually 'choke' the updrafts with showers quickly weakening but new cells developing nearby

@brucelehr95 @BethRigby Miatta Fanbulleh. it literally says it in the tweet

@BethRigby Who? What on earth kind of portfolio is that? Only labour could make up useless jobs like that. 🤣🙃

Breaking: first ministerial resignation- Minister for devolution faith and communities Miatta Fanbulleh has quit

The Starship transport barge "You'll Thank Me Later" has shown up in Florida with a brand new roof. 📷:NASASpaceflight

@AshleyKillip May 12th or 13th* launch

With lots of deluge tests at Pad 2 it looks like SpaceX Are trying hard to push for May 15th launch date

@natehiggins @limebike

Hi @limebike could you explain why you have blocked the entire central island with your bikes at this crossing on Westfield Avenue? Nobody can get through this - let alone anyone with a pushchair, wheelchair, with any other mobility issues or who is blind? Please get them moved!

@LadyJennifer23 @cloakzy It's Jen ceeeeena

EXCLUSIVE From Peanut's Wasteland @cloakzy finally answers the question everyone wants to know...

@lewisaoe stay umble

Official Pro Rankings 2026

@membtv ngl, too many aoe podcasts these days

I have make many podcast and interviews during my journey as streamer but i highly recommend this one. Inside the Mind of Liereyy: The Reality of the Redbull Wololo Londinium Final youtu.be/6qF_Ijj54WQ Hope you Enjoy.

@mattyxb seems does a lot except being a luggage suitcase and weighing a lot

I 3D printed my own suitcase to try and make traveling a little more tolerable.

🎶 Red Bull Wololo Londinium Sea Shanty 🎶 You’ve asked for it, here it is, enjoy! @AgeOfEmpires

@Sebastian_AOE @HeartttAoE What a match what a a tournament and what a champion you are for your fans! Congrats

🏰 Sin dudas una de las batallas más duras que tuve contra un gran rival como @HeartttAoE . 🙌🏽 Quiero agradecer la energía que me transmitieron ya que realmente la sentí en todo momento. ✊🏼 ¡Se que cuento con ustedes mañana!¡VAMO ARRIBA! #game #aoe2 #gaming #redbullwololo

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.

@JimDelahunt @FIFAcom @FIFAWorldCup Bring back the Qatar WC

Now so utterly ridiculous that cost is irrelevant. £250 would be a shrug & "that's the way it is", but £500 would be a nonsense & every 00 and 000 over that is a criminal assault on the spirit of football. @FIFAcom @FIFAWorldCup and their collaborators are a disgusting disgrace.

@Ryanair Imagine being you

average Man City attendance

@BenCrellin Who uses a spreadsheet for a calendar

Here's my updated World Cup schedule spreadsheet. Sorry to the 2.5k people who bookmarked my last spreadsheet - I didn't notice that FIFA changed the kick-off times for Brazil vs Haiti and Türkiye v Paraguay a couple of days ago. There are now 17 different kick-off times! #gid=0" target="_blank" rel="nofollow noopener">docs.google.com/spreadsheets/d…

@Dave_AoE Different era

@NewStadiumsDZ جهلاء وين تروح ترباو شوية

Des supporters qui envahissent le terrain, je comprends tout à fait la France qui autorise de - en - voire plus du tout nos matchs. Si les Pays-Bas voient ça, ils vont encore davantage réduire le % de supporters algériens & les reléguer dans les tribunes supérieures.

@WorldBollard where is the fun in this

These guys mean business. #WorldBollardAssociation 😎

@KITKAT Did Iran do this

Regarding recent press coverage