Cheryl Hung

Cloud Native London February 2024: DEFRA, AWS, and TitanML

Three speakers from DEFRA, AWS, and TitanML joined us at the February Cloud Native London meetup at the Just Eat offices, as well as via the YouTube/Twitch livestreams.


Kicking off our evening was Josh Greencroft, from DEFRA, telling us that There’s No Such Thing As Not Innovative Enough. Unfortunately due to technical issues we weren’t able to record his talk and he did not leave any takeaways.

Next up we had Marco Ballerini and Federica Ciuffo from AWS telling us about Kubernetes Multi-Cluster Made Easy(starting at 4:14). They discussed:

  • How to connect Kubernetes cluster
  • Why you should or should not use a service mesh
  • How you can leverage Amazon VPC lattice to connect clusters

And wrapping up our evening was Fergus Finn from TitanML discussing Challenges in serving self-hosted Large Language Model (LLM) applications(starting at 56:46). Check out his takeaways below:

  • Serving self-hosted LLMs is easier than you might think
  • There are other options besides ChatGPT for real applications
  • Lessons we learnt in building our LLM inference server

And of course, our usual group photos!

Cloud Native London March

Our next meetup will be on Wednesday 6th March, when we’ll be joined by speakers from VirtusLab, Parsectix, and more! RSVP and save the date now!.

Stay safe, stay healthy, and see you next month!

Cheryl (@oicheryl)