Xavier Badosa

Web experiments since 1995

[Xavier Badosa @badosa]

Read and write @JSONstat v2.0 statistical data in .NET. This library fills the C# gap in the JSON-stat ecosystem and mirrors the read/navigation interface of the JavaScript Toolkit.

https://t.co/ir9PnMNDsv

#csharp #opendata #nso #cubes

[Xavier Badosa @badosa]

New JSON-stat (read & write) library for the Go language. Programming languages currently supported in the @JSONstat ecosystem: #JavaScript, #Python, #Java, #Julia, #PHP, #R, #WebAssembly, #Rust & #Go.

https://t.co/fvhRR4idE1

#stats #cube #opendata #nso

[Andrej Karpathy @karpathy]

One pattern I find useful for working with LLMs is a nice long ramble session. Sometimes the LLM needs more bits to understand what you're trying to achieve, but you're too lazy to type them. In these cases I like to lean back, switch to /voice and just ramble for like 10 minutes, total mess, anything goes, full stream of consciousness. Sometimes I declare it up top, something like "switching to speech recognition sorry for any typos...". Sometimes I turn it into a small interview of a few turns. But I find that the LLMs are somehow very good at reconstructing long incoherent rambles and often their echo of your own tangle of thoughts comes out quite a bit cleaner than what you started with. The result is that you improve the mind meld and have to correct things less from that point on.

[Xavier Badosa @badosa]

How to import a CSV-stat (CSV-based format that, unlike similar solutions -Frictionless Data Package or CSVW-, requires a single file) into #Excel

https://t.co/KcVkPlpZBH

You only need to create an Excel Workbook with a set of fixed #PowerQueries (#M). #opendata #cube #stats

[NXT EU @NXT4EU]

Germany has launched one of the world's best open-source AI models.

Soofi S, made by the Soofi consortium, is a 30B parameter model fully trained in Europe and tops the ranking for open-source AI.

Huge moment for Europe, and finally some competition for Chinese open-source AI.

[Mira Murati @miramurati]

Our first model, Inkling. Trained from scratch, weights are open, fine-tunable on Tinker today.

[Xavier Badosa @badosa]

Bidirectional bridge between the columnar stack (Apache #Arrow, #Parquet, #DuckDB, #Polars), several CSV-based formats (plain #CSV, Frictionless Data Package, CSV-stat, #CSVW) and @JSONstat 2.0 datasets.

https://t.co/kJSt5TN9Xa

#cube #opendata #stats #nso

[JSON-stat.org @jsonstat]

JSON-stat Builder is a browser-based tool for creating @JSONstat interactively. You describe your data through a friendly step-by-step wizard and get back a valid, ready-to-publish dataset.

https://t.co/0xxhCrrzp7

#builder #cubes #opendata #stats

[Xavier Badosa @badosa]

jsonstat-validator is a validator for @JSONstat. It reuses the JSON Schema defs for the structural pass & layers on top a rules engine that checks the cross-field cube invariants https://t.co/AN61K89bRn
#TypeScript #Rust #WebAssembly #CLI #NodeJS #browser

[EDU @edusoyy]

🚨ANTHROPIC ACABA DE PUBLICAR EL MANUAL PARA MONTAR UNA EMPRESA SIN EMPLEADOS

>CEO: 1 persona.
>empleados: agentes de claude

dura 30 minutos y es gratis

Guarda esto en favoritos para que no lo pierdas.

[Xavier Badosa @badosa]

https://t.co/iyAGEB1JMT

#Open Knowledge #Format “represents catalog knowledge as plain #markdown files with #YAML frontmatter, organized in a directory hierarchy. That choice unlocks a few properties that are hard to get from a service-owned metadata store.”

[Xavier Badosa @badosa]

jsonstat-wasm: #Rust and #WebAssembly packages for @jsonstat

Now you can use Rust and WebAssembly to read, traverse and transform JSON-stat responses using a subset of the jsonstat-toolkit API ("the good parts").

https://t.co/9PeqNoWUc6
https://t.co/oxYgiA10MZ

[JSON-stat.org @jsonstat]

A new data explorer for JSON-stat has been released: the JSON-stat WebMCP Explorer. The defining feature of this viewer is its support for the emerging #WebMCP standard (currently under development).

https://t.co/jziygMSfXV

https://t.co/szLBSUYO9W

[Idescat @idescat]

L'Idescat intervé en la reunió plenària del projecte “AI-Ready Dissemination”, organitzada per la Comissió Econòmica de les Nacions Unides per a Europa, que se celebra avui en línia
@badosa presenta: “LLMs & the WWW: Towards the Agentic Web”
👉 https://t.co/CWYp2bbA6s

[Movez @0xMovez]

Andrej Karpathy:

“Vibe coding is incredible. But agentic engineering is the next level.

90% of my coding routine is automated by AI agents.”

In this 30-minute talk, Andrej Karpathy explains how to build an AI agent workflow from scratch.

Worth more than 500$ agentic engineering course on the internet.

Watch it today, then read the article below.

[JSON-stat.org @jsonstat]

The initial release of an LLM-maintainable knowledge base for the JSON-stat format is now available.

https://t.co/cw2I8zdsnZ

This repository adheres to the LLM Wiki pattern (as popularized by #Karpathy) to ensure seamless integration with modern AI workflows. #llm #wiki #stats

[Andrej Karpathy @karpathy]

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.

[Xavier Badosa @badosa]

#Chess #Clock with support for #Fischer increment and #Bronstein delay, the latest mini-app at https://t.co/h2WRmhgRQZ (see also the Chess Image Generator)

https://t.co/zTvH2AcpUY

[JSON-stat.org @jsonstat]

Learning @jsonstat ? New repository of JSON-stat processing samples https://t.co/MlTTQ4JQi4 #Dataviz of data from providers around the globe like @EU_Eurostat @UNECE @ONS Statistics Norway @CSOIreland @StatsFinland @StatDataCH @es_INE @idescat @IGE_Estatistica Statistics Denmark

[Xavier Badosa @badosa]

https://t.co/h2WRmhgRQZ, a free repository for mini-apps mostly created for my kids: ads-free & safe tools and non-epileptic games. #Math escape room, #chess image and FEN notation generator, #sudoku, #reversi, #midi generator...