iiinova — where your digital self keeps evolving

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 05/08/2026

Attention Berlin people: On August 10 (next Monday), I'm co-hosting a meetup with Hugo and Jodie @t-redactyl.bsky.social at @jetbrains.com 🚀 Featuring: agents, agentic NLP, coding assistants and lightning talks. See you there! 📆 Mon, Aug 10, 6–9 pm 📍 JetBrains, Berlin ✅ RSVP: luma.com/agmd8ie4

1114

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 27/07/2026

I gave a talk on "Build-to-own AI: Agentic Development for Humans" at PyCon Armenia: how coding agents change the way we build ML systems and why code and programmable interfaces matter more than ever, not less. Here's the whole thing as a thread 🧵 (1/18)

1155

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 23/07/2026

In Yerevan, Armenia, for PyCon Armenia & PyData Yerevan! I'm giving the opening keynote on Friday (tomorrow) and will be talking about "build-to-own AI", agentic development for humans, and how to use AI to build better AI. Looking forward to seeing everyone and exploring Yerevan on my day off ✨

052

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 04/07/2026

📣 Attention Prodigy users: If you're already using Prodigy for annotation, we've put together a page on how our new tool and platform Ellf can help you scale up and automate your data development projects. beta.ellf.ai/prodigy

9252

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 11/06/2026

Some thoughts on how to do beta testing well, the importance of building things and using your own tools, and what we're currently doing:

We're fully back in "building new things" mode and have been thinking a lot about how to do beta testing for Ellf, our new tool for agentic NLP development, and how to make sure we're shipping updates quickly. One thing we've found from our other developer tools like spaCy and Prodigy is that it's crucial to be users of our tools as well. We need to be solving real problems and ship real projects ourselves.

Otherwise, it's way too easy to work in a bubble and build things that aren't actually useful in practice. So many features and improvements were direct results of us building things and stumbling over missing functionality, unideal UX or straight-up bugs. ("Why is this so bad, who wrote this? Oh... me.")

Ellf has a lot of moving parts: the coding assistant integration for project planning and NLP advice, the collaborative platform for annotation, batch jobs and agents, and the data-private cluster hosted by the user to run things and manage data. Ideally, we need to be doing and using all of it, and run as many end-to-end workflows as possible: plan a pipeline or component, create data, automate with agents, train a model, evaluate, iterate and collaborate.

So here's what we're currently doing:

With our beta partners, we've started scoping out practical NLP problems and are solving them at a big discount, while collaborating on development via the platform, and comparing Ellf's project advice to our solution. In the end, we get to ship real and fully independent deliverables – like a custom component, pipeline or workflow. To make sure we're getting to do many different projects, we're targeting very quick turnaround times of a few weeks and very clearly scoped deliverables.

We're especially interested in use cases like information extraction and structured data, turning LLM prototypes into modular pipelines that are better, faster, cheaper and private, human-in-the-loop distillation and of course data development.

So if you and your team need help with a burning NLP problem or use case and want to try out "Vibe NLP" and agentic NLP development, feel free to get in touch!

682

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 14/05/2026

Want to be a beta partner of our upcoming product for agentic NLP? We're looking for teams who need help with their NLP projects, especially around tasks like information extraction. Ellf is a virtual assistant for NLP projects – think Claude Code, but with NLP superpowers, or "Vibe NLP" ✨

1218

Matthew Honnibal @honnibal.bsky.social · 24/03/2026

Open-source projects should stop making the public project repo a Pypi trusted publisher. Configure the public repo's CI to make a release with all the wheels and other release artefacts attached. Then configure a private repo to do the publication. GHA is really easy to mess up and get pwned.

161

Matthew Honnibal @honnibal.bsky.social · 06/03/2026

What we recognise as emotions in ourselves and non-human animals requires a pretty specific interaction between cognition, the body and reward/aversion. There's no reason why exactly the weights should happen to configure themselves into an exact analogue of that thing.

100

Matthew Honnibal @honnibal.bsky.social · 01/03/2026

A self-replicating thing doesn’t have to be very smart to cause major problems. Generally we can plan ahead though, and contain the damage. Well, we could do that. In theory. Or we could spice things up a bit. Maybe run some bat-licking ecotours instead. Why not? honnibal.dev/blog/clownpo...

honnibal.dev

The looming AI clownpocalypse · honnibal.dev

Exploits will soon be cheaper to develop autonomously than they earn. What then?

0103

Matthew Honnibal @honnibal.bsky.social · 19/02/2026

When I'm programming, the text I get from Claude Code after it's done some analysis is not at all bad. But if you tell it to write something, there's nothing it's "trying" to say. If you tell it "make words exist" it does, but then you look at them and go "this sucks actually". And it does.

020

Matthew Honnibal @honnibal.bsky.social · 17/02/2026

You'll laugh but I do get it to test the tests sometimes actually, with this sort of thing: github.com/honnibal/cla... . After it generates the list, I ask another session to introduce the bugs, and check whether the current tests change them.

github.com

000

Matthew Honnibal @honnibal.bsky.social · 16/02/2026

I guess my socials are way more tech focussed than my...social, so maybe this take gets kind of a "duh" here. Not sure. But: Why I don't think AI is a bubble honnibal.dev/blog/ai-bubble (basically an argument that LLM progress probably won't plateau. Not a finance guy)

honnibal.dev

Why I don't think AI is a bubble · honnibal.dev

Matthew Honnibal is a computational linguist from Sydney based in Berlin, Germany. He's the author of the spaCy Natural Language Processing library and the co-founder of Explosion.

2143

Matthew Honnibal @honnibal.bsky.social · 15/02/2026

Basic take I know, but "the information stays on ChatGPT" and "the information is used to decide which ads to serve" are not compatible statements.

120

Matthew Honnibal @honnibal.bsky.social · 08/02/2026

If your tests ever find anything interesting, Claude Code will challenge you to a little contest-of-wills where it tries to cope, rationalise or lie egregiously to avoid fixing the problem. I call this minigame 'sneaky quokka'. honnibal.dev/blog/sneaky-...

honnibal.dev

Sneaky quokka: Testing and debugging with LLMs · honnibal.dev

Matthew Honnibal is a computational linguist from Sydney based in Berlin, Germany. He's the author of the spaCy Natural Language Processing library and the co-founder of Explosion.

061

Matthew Honnibal @honnibal.bsky.social · 04/02/2026

Yeah that's definitely a related problem. If you use an LLM to generate a web service, it'll try/except everything and just log an error. If you have something that shouldn't possibly occur it's better to have a stack trace

000

Matthew Honnibal @honnibal.bsky.social · 04/02/2026

The gist of it is that it's really easy to write subtle bugs with try/except in Python, and there's a history of bad advice encouraging these bugs. If you're using an LLM coding assistant you're going to need to police its try/except usage carefully, so it's a good time to finally clear this up

020

Matthew Honnibal @honnibal.bsky.social · 04/02/2026

I've been working on a post on LLM-assisted coding, and I guess I've built up a bunch of writer's debt, because it turns out there's a bunch of tangents I wanted to write up first. One really concrete thing that came up is try/except usage -- so, here's the post. honnibal.dev/blog/try-exc...

honnibal.dev

Exception handlers are not general-purpose conditionals · honnibal.dev

Matthew Honnibal is a computational linguist from Sydney based in Berlin, Germany. He's the author of the spaCy Natural Language Processing library and the co-founder of Explosion.

2122

Matthew Honnibal @honnibal.bsky.social · 10/06/2025

It's so hard to tell though, because the context of every task is so different. I feel like I'm accumulating superstitions about what helps and what harms.

040

Matthew Honnibal @honnibal.bsky.social · 10/06/2025

Now it feels like sycophancy has skyrocketed? It tells me how 100% right I am every prompt, and it gives me confidently wrong "Job done boss!" assessments like "The tests are all passing" when they're just not. Have I made it...insecure? Like did I prime a unhappy boss/nervous worker social script?

110

Matthew Honnibal @honnibal.bsky.social · 10/06/2025

I feel like my Claude Code got a lot worse after I made some grumpy changes to my system prompt. I was annoyed that it kept trying to condition around invalid data instead of failing and fixing the data at the source, so I added some instructions to the prompt that were maybe too blunt. However...

240

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 09/06/2025

I'm back from a busy four days in London! 💜 I had a great time at @pydatalondon.bsky.social‬ and you can read my recap and highlights here: www.linkedin.com/feed/update/...

0174

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 02/06/2025

Hi Berlin people! @hugobowne.bsky.social is in town & we're celebrating by hosting a meetup together 🎉 This one is all about building with AI & we'll also open the floor for lightning talks. If you're around, come hang out with us! 📆 June 16, 18:00 📍 Native Instruments (Kreuzberg) 🎟️ lu.ma/d53y9p2u

084

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 26/04/2025

Another great @pyconde.bsky.social is over and I'm exhausted but happy, and feel very inspired! ⚡ Here are my personal highlights: www.linkedin.com/feed/update/...

1155

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 05/04/2025

My friend @hugobowne.bsky.social launched new course on Building LLM Applications for Data Scientists and Software Engineers! I'll be joining for a guest lecture on human-in-the-loop development and distillation workflows. You can get 10% off with my code + lots of perks: maven.com/hugo-stefan/...

1223

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 31/03/2025

I don't normally write advice columns but this question I got deserved its own post: With all the hype around Generative AI, how can you, as a developer, communicate the nuances and advocate for new and modular solutions that are better, easier and cheaper? explosion.ai/blog/modular...

3258

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 29/03/2025

So many good events coming up this season. Here's my schedule – hope to see you there! Apr 10: data:unplugged (Münster) Apr 23: @pyconde.bsky.social (Darmstadt) May 31: @pyconit.bsky.social (Bologna) Jun 5: ECONDAT (London) Jun 7-8: @pydatalondon.bsky.social (London) 📆 Details: explosion.ai/events

094

Reposted by Matthew Honnibal

Florentine Dömges @fdoemges.bsky.social · 13/03/2025

@explosion-ai.bsky.social Only the best companies will send you stickers if you are a fan and beg for them! Thank you so much, Ines and Matt!!!!

033

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 24/03/2025

New case study: How Love Without Sound built innovative AI-powered tools for the music industry and law firms, and helped publishers recover hundreds of millions of dollars in lost revenue for artists. explosion.ai/blog/love-wi...

1175

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 25/02/2025

Out now: Prodigy v1.18! New features include: ✍️ edit text during span annotation 💬 fully custom UI translations 🎛️ new JavaScript events and functions ✅ various small fixes and docs More details: prodi.gy/docs/changelog

0123

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 14/02/2025

These are the kinds of NLP stories I love! TFW you "accidentally" train a great 2 MB (!) task-specific model 🤯 @strickvl.bsky.social

1334

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 13/02/2025

Look what arrived in the mail today! 🎉 This is the 2nd edition of "Mastering spaCy" by Duygu Altinok and Déborah Mesquita, featuring how to build structured NLP solutions with custom components, and updated content on using models powered by LLMs. You can get it here: www.amazon.com/dp/B0DVBTX2BL

1416

Matthew Honnibal @honnibal.bsky.social · 06/02/2025

spaCy development live stream continues today! Today I'll be looking at getting Confection updated, to allow Python 3.13 support and add a file-management feature. Playlist: www.youtube.com/watch?v=TTEW... The spaCy+PyTorch design has been going well, but we need that file management feature now.

youtube.com

Streaming spaCy (Feb 6, 2025): Updating Confection for Python 3.13 and more

YouTube video by Explosion

061

Matthew Honnibal @honnibal.bsky.social · 31/01/2025

Pretty happy with how the spaCy+PyTorch design is coming together. And hopefully I'll continue to get smoother at streaming :) www.youtube.com/watch?v=YeKi...

youtube.com

Streaming spaCy (Jan 31, 2025): Continue exploring PyTorch+spaCy

YouTube video by Explosion

020

Matthew Honnibal @honnibal.bsky.social · 30/01/2025

Live again now, doing more design work on what it would look like if we rebuilt spaCy on PyTorch: www.youtube.com/watch?v=fzFH...

youtube.com

Streaming spaCy (Jan 30, 2025): Continue exploring PyTorch+spaCy

YouTube video by Explosion

040

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 27/01/2025

Enjoyed giving my keynote on "What the history of the web can teach us about the future of AI" at PyCon+Web. I wrote it up as a blog post because I think there are many interesting parallels and lessons we can learn: explosion.ai/blog/history... Here are the most important points 🧵

explosion.ai

What the history of the web can teach us about the future of AI · Explosion

How will AI development look in the future? There is a lot we can learn from another groundbreaking technology: the web. This blog post takes a look at what the history of the web can teach us, and wh...

1336

Matthew Honnibal @honnibal.bsky.social · 17/01/2025

Okay let's try this again -- Live now! Streaming software issues hopefully sorted 🤞

141

Matthew Honnibal @honnibal.bsky.social · 16/01/2025

More OBS problems. Will try to get live when I've sorted it out 🤞

000

Matthew Honnibal @honnibal.bsky.social · 16/01/2025

To explore solutions, I'm going to start from a sandbox. Imagine we were creating this new library. Call it "spycy". With no attention to spaCy backwards compatibility, what would we do? If we can get a sandbox solution we like, we can then look at the least disruptive way to adapt spaCy.

000

Matthew Honnibal @honnibal.bsky.social · 16/01/2025

In the stream I'll talk through why supporting PyTorch well has been hard. The gist of it is that the Language class that owns the pipeline is too involved in the optimisation details. This prevents us from using other libraries' high-level training abstractions.

100

Matthew Honnibal @honnibal.bsky.social · 16/01/2025

The BLIS compilation problems and numpy compatibility mess I've dealt with over the last couple of weeks have put new focus on some longstanding issues for spaCy. Our GPU support still sucks, and we have this numeric computing stack that's not working for us. How can we switch to PyTorch?

100

Matthew Honnibal @honnibal.bsky.social · 16/01/2025

Will do another short (~2h or so) stream this afternoon, at 13:00 CET. Last week I was limited in what I could do because the main thing I was working on was build system stuff, and OBS was crashing when I tried to create a desktop view. This week we can look at a more interesting topic.

2102

Matthew Honnibal @honnibal.bsky.social · 09/01/2025

Going live now! Join me for the first pilot stream on YouTube: www.youtube.com/live/kViiI5B...

youtube.com

YouTube

Share your videos with friends, family, and the world

083

Matthew Honnibal @honnibal.bsky.social · 06/01/2025

Now that Explosion is small again, I'm also missing a bit of interactivity in my work. Looking forward to seeing how this goes! I want to do the first pilot streams this Thursday 2pm CET (8am EST) and Friday at 10am CET on our YouTube channel. youtube.com/c/ExplosionAI

youtube.com

Explosion

Explosion is a software company specializing in developer tools and tailored solutions for Artificial Intelligence and Natural Language Processing. We’re the makers of spaCy, one of the leading open-s...

060

Matthew Honnibal @honnibal.bsky.social · 06/01/2025

Trying something new this year: I'm going to stream spaCy and other open-source development and maintenance twice a week! I hope this will give the community a better inside look while things are in progress. I'll also answer your questions and chat about NLP, OSS and more.

822813

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 30/12/2024

Writing a new talk on "What the history of the web can teach us about the future of AI" 🔮 I've wanted to do this for so long because I think there are some great lessons and analogies here. I'll be presenting it for my keynote at PyCon+Web in Berlin on Jan 25 – hope to see you there!

1212

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 14/12/2024

Happy Birthday to me! 🥳 It feels like a good opportunity to look back at 2024, an eventful but also difficult year for me. So here's my personal review, including travel, talks, writing and various things I did and enjoyed. ines.io/blog/year-in...

ines.io

My Year in Review 2024 · ines.io

2024 was an eventful but also difficult year for me. This is my personal summary, including travel, talks, writing and various things I did and enjoyed.

3353

Reposted by Matthew Honnibal

Ines Montani 〰️ @inesmontani.bsky.social · 02/12/2024

New post: From PDFs to AI-ready structured data 📃✨ A deep dive into document processing, layout analysis and a modular workflow for building end-to-end document understanding and information extraction pipelines using PDFs, Word documents, scans and more. explosion.ai/blog/pdfs-nl...

66611