TinyML: Bringing machine learning to the edge

Thoughtworks Technology Podcast

Thoughtworks

Careers, Business, 907234, Technology

4.7 • 53 Ratings

🗓️ 4 May 2023

⏱️ 46 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

When we think about machine learning today we often think in terms of immense scale — large language models that require huge amounts of computational power, for example. But one of the most interesting innovations in machine learning right now is actually happening on a really small scale.

Thanks to TinyML, models can now be run on small devices at the edge of a network. This has significant implications for the future of many different fields, from automated vehicles to security and privacy.

In this episode of the Technology Podcast, hosts Scott Shaw and Rebecca Parsons are joined by Andy Nolan, Director of Emerging Technology at Thoughtworks Australia, and Matt Kelcey of Edge Impulse, to discuss what TinyML means for our understanding of machine learning as a discipline and how it could help drive innovation in the years to come.

Transcript

Click on a timestamp to play from that location

0:00.0	So welcome, everybody. My name is Scott Shaw. I'm one of the hosts of the ThoughtWorks
0:14.3	Technology podcast here. And the topic today is going to be TinyML.
0:23.5	And we'll find out what that is in a second.
0:29.2	But there are the three other people here that would like to introduce themselves.
0:30.9	So Rebecca, do you want to go first?
0:32.3	Sure.
0:34.0	This is Rebecca Parsons.
0:38.7	I'm another one of the recurring hosts for the ThoughtWorks Technology podcast,
0:48.9	and I'm here to be the peanut gallery. Andy, do you want to go next? Yes, I'll go next. So I'm Andy Nolan, Director of Emerging Technologies at ThoughtWorks Australia. Really excited to talk about TinyML today.
0:56.2	And Matt, would you like to go next?
0:58.7	Yes, hello. My name's Matt.
1:00.8	I'm a software engineer machine learning person at a tiny ML startup called Edge Impulse,
1:06.6	so long-term ML person and more recently focused on the smallest of small email.
1:13.0	And Matt used to work with us at ThoughtWorks and we were sorry to lose him but we were also
1:18.7	excited about his new venture.
1:20.3	And before that he was at Google doing some of the work there that's gotten some renown. Let's start out with some definitions.
1:31.9	Andy, you're the first person that brought this to my awareness, I guess. I wonder if you could
1:37.9	give us a bit of an overview of what are we talking about with TinyML? Yeah, for sure. So TinyML, the name says a lot.
1:45.7	So it's a very small version of what we typically know as machine learning.
1:50.3	And it focuses on deploying models to sometimes the smallest device as possible.
1:55.8	So I think microcontrollers, Raspberry Pi's, those types of devices.
2:01.0	And it does have its constraints around what you can do with it, but it also has a whole
	...

Transcript will be available on the free plan in -695 days. Upgrade to see the full transcript now.

Previous episode | Next episode

Disclaimer: The podcast and artwork embedded on this page are from Thoughtworks, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of Thoughtworks and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.