How @camunicampop.bsky.social researchers are turning static snapshots of Victorian Britain into dynamic life histories. The team are linking over 200 million census records, revealing how people moved, worked and lived in ways never before possible. 🔎 Read on 👉 www.hist.cam.ac.uk/news/tracing...

Continuing on my search for good game mechanics for maths games for kids. This one is a lane crossing game: rupertlinacre.com/number_lanes/ This seems to work well because it's fun (at least accordingly to my 7yo), but there's no time pressure to answer the questions.

Introducing letterpaths 1.0, free software for teaching cursive writing. Blog w/ examples, videos + worksheet generator: www.robinlinacre.com/letterpaths_... As far as I can tell, this is the best free and open source library for building handwriting tools according the the UK curriculum

As capabilities have improved, I am increasingly sceptical of the description of LLMs as a 'calculator for words', and the idea that they excel at semantics, because their rate of improvement in these areas seems extremely modest in comparison to things like coding and tool use

Here's an example of how my `letterpaths` library can be used to power a game, intended for phones/other touchscreen devices Game: www.robinlinacre.com/letterpaths/... Code: github.com/RobinL/lette... The game can be installed as an android app so it can be used fullscreen. It's free+open source

Developing a cursive writing typescript library capable of powering educational apps. It's free an open source and has: - Geometry for each letter - Algorithm for joining letters - Algorithm for animating cursive writing - Algorithm for tracing/path finding Link in 🧵

Finding working with agents all day a bit mind bending. Always having to think about how to enable an agent to do good work, and how it verifies its work.

After a period of stagnation in UK solar nice to records starting to be broken again in 2025 and now 2026 by significant margins

The country quiz now has: - Capital cities mode - Single continent mode - 'What country is highlighted' mode (following a travelling salesman route) - More detailed borders - 4k mode, with much bigger map - Choice of map projection rupertlinacre.com/country_quiz/ Access via options menu.

Reading The Baroque Cycle - King of the Vagabonds (2003). This quote hit me: “What does the Doctor want?” “To translate all human knowledge into a new philosophical language, consisting of numbers. To write it down in a vast Encyclopedia that will be a sort of machine...

In the next five years, we are going to do LOTS of census linking 🔗🔗🔗🔗🔗, produce new transcriptions of all UK censuses 1841-1921, and use these to study migrations and demographic transitions, thanks to @ukri.org ! Sneak peek at: www.geolinking.hist.cam.ac.uk

uk_address_matcher v1.0.1 is released: - Hugely reduced memory usage, especially when matching to full UK. - Over 30% faster - Now supports duckdb 1.4.4 and 1.5.0 - New docs on choosing a matching threshold and optimising accuracy here: moj-analytical-services.github.io/uk_address_m...

I have been enjoying playing Sporcle's countries of the world quiz with my son, so I made a fancier version. Play here: rupertlinacre.com/country_quiz/

We are pleased to release `uk_address_matcher`, a free Python package for address matching and geocoding, developed by Tom Hepworth and me. The package has several aims: simplicity, speed and accuracy.

📣 NEW! I’ve just released the BIGGEST and perhaps most creative project I’ve ever worked on! “Searching for Birds” searchingforbirds.visualcinnamon.com 🐤 A project, an article, an exploration that dives into the data that connects humans with birds, by looking at how we search for birds.

In case you missed it: New blog: Respectful use of AI in software development teams www.robinlinacre.com/respectful_u... LLMs are increasingly able to write production quality code. But what cognitive work can be delegated to LLMs without damaging the health of the team?

For anyone with FOMO wondering whether to pay for Opus 4.5/Claude Code, my experience is that OpenAI Codex is very similar in performance. i.e. both are excellent, but Claude Code is not a magic unlock

Excited to be a part of the initiative to "Move Fast and Fix Things", announced in Chief Secretary to the Prime Minister's speech today. One measure is an expansion of the No10 Innovation Fellowship, for which we've launched a new website! fellows.ai.gov.uk Speech: www.gov.uk/government/s...

New blog: Respectful use of AI in software development teams robinlinacre.com/respectful_u... LLMs are increasingly able to write production quality code. But what cognitive work can be delegated to LLMs without damaging the health of the team?

Made a simple game to help my daughter learn her alphabet. Took about 1 day for full code, >200 images >200 voice files and music nano-banana-pro is incredible. But it's also so quick to vibe code image and audio processing scripts. robinlinacre.com/bee_letters/ github.com/RobinL/bee_l...

I struggle to find no-nonsense, free and 'fun'(ish) maths games for my son (7yo) so I have been making a few Here's another one: Maths vs monsters. This is his fav so far rupertlinacre.com/maths_vs_mon... Code: github.com/rupertLinacr... Other games/maths utilities: rupertlinacre.com

OpenUK Awards 25 Open Data Category sponsored by Open Data Institute, Shortlist is live, congratulations to the shortlisted nominees: Ministry of Justice UK Splink Team (@robinlinacre.bsky.social), OpenActive, and UK Power Networks (Yiu-Shing Pang) 🍾🥂🏆 #openukawards #opensource #opendata

More progress on #openaddresses: Islington Council in London has released its Council Tax address list for re-use as #opendata under the Open Government Licence www.owenboswarva.com/blog/post-ad... I've made a geocoded version by adding coordinates from ONS #FOI #localgov #UKhousing #proptech

New ✨interactive✨ explainer: Address matching using a fault tolerant trie: robinlinacre.com/fault_tolera... Which illustrates a powerful technique for address matching that we're currently working on building into uk_address_matcher (github.com/moj-analytic...)

When working a complex postgres schema, I find it time consuming to figure out the joins. I had an idea: a 'join generator' that traverses the relationship graph for you, and writes the joins. You give it a dump of the postgres schema, and it gives you a UI. www.robinlinacre.com/vite_live_pg...

We're working on a DuckDB community extension called `splink_udfs` to add some record linkage related functions to DuckDB. It's currently very much WIP, but you can already use it wherever you're using DuckDB. github.com/moj-analytic...

If you're using Splink with DuckDB you should see significant speed improvements by updating to DuckDB 1.3.x. You can also add more granularity to your comparison levels statements without an impact on run times. Depending on your model spec, it could be twice as fast or better.

My most commonly used pattern for AI coding: Dump entire source code into Gemini 2.5 pro, write prompt specifying what I want, and then: Give precise instructions for an LLM to follow to implement this feature. Break the solution down into steps where each step is verifiable.

I have been working on a free, high performance address matcher. I've written up some key tricks, techniques, and ideas into a blog post here: www.robinlinacre.com/address_matc...

The 'build' button in google AI studio is unbelievably good. I had an idea to visualise fractions, three prompts total and it's pretty close to something useful (it does this for any arbitrary fractions). Even the one-shot attempt was pretty good

My PyData Global talk "Rapid deduplication and fuzzy matching of large datasets using Splink" is now on Youtube: www.youtube.com/watch?v=eQtF...

I see too much focus on trying to find applications of LLMs to help other people 'at scale' with their jobs. At the moment, the output of LLMs is rarely useful for business rules or passive consumption. The lower hanging fruit is helping people use AI directly & however they see fit in their job.

If you're using duckdb in a python script or jupyter notebook, you can run con.execute('CALL start_ui()') at any point, and the ui will pop right up in your web browser with the current database automatically available. (I knew about the UI, but I had missed this trick!)

Gemini 2.5 pro is really good. Grok 3 felt like a big step forwards and was my 'go-to' for hard problems, and this feels like another significant step forward. So nice with small codebases to be able put everything into context (I use github.com/simonw/files... )

Ended up writing a follow up post with the final approach and learnings from getting this running on GitHub Actions! All original datasets weight more than 500GB combined. The final ones published on 🤗, only 1 GB. Took some tinkering to get there but was fun! davidgasquez.com/exporting-in...

New blog: Why DuckDB is my first choice for data processing: www.robinlinacre.com/recommend_du...

I vibe coded a primary school maths breakout game - aimed to be fun and educational. rupertlinacre.com/breakout_mat... In the process I created and open sourced a maths problem generator aligned to the national curriculum, so you can vibe code your own maths games! www.npmjs.com/package/math...

Just added an example/tutorial to the Splink docs of matching business data. It uses some feature engineering tricks that help improve accuracy vs. just fuzzy matching on names. moj-analytical-services.github.io/splink/demos...

I think the single most productivity-enhancing use of LLMs in gov would be give all devs and data scientists access to Cursor (or equivalent). I am not yet convinced of the widespread value of 'behind the scenes' uses of LLMs, but v. bullish on skilled human-in-the-loop uses, especially coding

With DuckDB WASM it's possible to run a full Splink model in your browser in a single, standalone .html page. Here's an example: www.robinlinacre.com/live_splink/ And the git repo: github.com/RobinL/vite_...

Playing around with a spatial duckdb wasm database in a static webpage. Absolutely amazing how far you can get with geospatial in the browser using entirely open source tools

Writing code with LLMs: 1. Iteratively try to explain what you want using natural language until the code works 2. You now have a precise prompt (the working code). Use it to get high quality code.

Splink has just passed 10 million downloads on PyPi, the first UK gov package to do so! If you're a user, we'd be super grateful if you could support Splink by adding your org to our list of users: moj-analytical-services.github.io/splink/index.... To do so, please get in touch, or raise a PR.

New blog: AI probably won't replace me in 2025 www.robinlinacre.com/llms_in_2025/

Advanced Voice Mode + video as a reading companion is incredible as an educational tool. Completely transformational for teaching/learning something like Shakespeare. It can read aloud, translate into simpler/modern English, explain what's going on etc.

New blog: The emerging impact of LLMs on my productivity In which I take a no-hype look at some ways LLMs have been helping me. And make some (probably inadvisable) predictions about what we might see next. www.robinlinacre.com/two_years_of...

LLM context that summarises the Splink docs + probabilistic linkage theory to add to your prompts: moj-analytical-services.github.io/splink/topic... Long = 83k tokens, short = 25k tokens. Seem to work pretty well 👍 You can use this to write Splink code, explain concepts or code, and so on

Loved this post from @robinlinacre.bsky.social! www.robinlinacre.com/parquet_api/

I'm presenting a 30 min introduction to Splink at PyData Global - 4th December, 16:30–17:00 (UTC), Tough act to follow though 😅 , hope I can make it worth sticking around! Link to my talk 👇 global2024.pydata.org/cfp/talk/YXU...