Topic: Software
by Aravind M
This week, we talk about Natural Language Processing and machine learning with our special guest, Ines Montani.
Ines is the co-founder and CEO of Explosion.
spaCy: https://spacy.io
Prodigy: - https://prodigy.ai
Prodigy Teams (our upcoming product): https://prodigy.ai/teams
spaCy's config system powered by Pydantic:ÃÂÃÂÃÂÃÂÃÂÃÂÃÂ https://explosion.ai/blog/spacy-design-concepts
Textual UI for Prodigy by Vincent Warmerdam: https://github.com/koaning/prodigy-tui
Large Language Models: From Prototype to Production (EuroPython keynote):
The Guardian case study: https://explosion.ai/blog/guardian
Using LLMs for more efficient annotation: https://prodi.gy/features/large-language-models
spacy-llm Integrating LLMs into structured NLP pipelines:ÃÂÃÂÃÂÃÂÃÂÃÂ https://github.com/explosion/spacy-llm
Chapters:
[00:00:02] Mike Driscoll welcomes Ines Montani to the Python show
Ines Montani is the co founder and CEO of explosion. She also works as a developer on open source projects.
[00:01:45] How do you found a company around an open source software package?
I've always programmed. I started as a teenager on the Internet. My first product was an annotation tool that was also scriptable in Python. How do you found a company around an open source software package?
[00:06:47] You've been going around the Python conference circuit this year
It's really exciting again to be at real conferences. We also did our first conference booth, first ever at PI data Amsterdam. Still have more coming up this year. I love the python community.
[00:07:30] You spoke at some conferences about how LLMs fit into a practical workflow
I saw you were speaking at some of those conferences. How did that go? Yeah, really, really good. Everything was recorded. I'll try to include. for your show notes. I think Europython is now online and some other conferences are following.
[00:08:58] What are your favorite python packages that aren't ones that you've helped create?
What are your favorite python packages that aren't ones that you've helped create? I've always had a soft spot for making things look nice, both on the web and in the terminal. Have you checked out the textural project yet?
[00:11:12] How did your company Explosion come about?
In the beginning, we were actually quite small. And then slowly, one, two, three more people. And we also started out profitable because we saw, like, hey, actually, what? Also, it's really important to start out early, running a business.
[00:12:34] spaCy is a natural language processing library that can be used in production
Spacy is a library that can be used in production. It's built on the idea of you want to process text. There are lots of different use cases from all kinds of industries. Prodigy has always focused on data privacy and allowing users to run things on their own hardware.
[00:19:12] What are some of the most notable applications that were created using spaCy?
So what are some of the most notable applications that were created using Spacy or prodigy? And then there's some other ones that are notable in the sense that they provide an insane amount of value, but are internal and kind of unsexy.
[00:21:52] Getting into programming and getting into development has been easier than ever before
Getting into programming and getting into development has been easier than ever before with models that can help you program. And then on the other side, of course, we also see a lot of new capabilities that we can use and integrate. How do you think all these new LLMs will affect your company?
[00:28:31] What currently excites you about the machine learning space right now
What excites you about the machine learning space right now? Can we use a large language model to help people be better at applying best practices when they're building machine learning systems? Open source is still going to, going to win and become or be important.
[00:32:27] Thank you so much for being on my show today
I hope we can find some time to hang out again. You can tell me more about LLMs and increase my knowledge on machine learning.