If you're keen to share your story, please reach out to us!

Guest:

https://www.linkedin.com/in/agnes-leroy/

https://jobs.zama.ai/

Powered by Artifeks!

https://www.linkedin.com/company/artifeksrecruitment

https://www.artifeks.co.uk

https://www.linkedin.com/in/agilerecruiter

LinkedIn: https://www.linkedin.com/company/enginearsio

Twitter: https://x.com/Enginearsio

All Podcast Platforms: https://smartlink.ausha.co/enginears

00:00 - Enginears Intro.

00:48 - Agnès & Zama Intro.

01:50 - The journey of Zama from the start.

03:17 - How is Fully Homomorphic Encryption (FHE) different from traditional encryption methods?

04:26 - What made FHE possible in 2009?

05:41 - What is Zama's mission and where is Zama's FHE being utilised?

06:53 - Where is FHE being applied in the market today and why?

10:25 - Where are the biggest challenges in protecting data at the moment?

11:35 - How do you make the computations x100 faster?

12:59 - Where are the performance challenges for AI?

14:07 - Why is it so costly to run and train LLMs?

15:11 - How & why are Zama using Rust + other technologies?

16:58 - Decisions when choosing what technologies to use - how to go about choosing?

18:54 - Problems Agnès & the Zama GPU team are solving today.

21:24 - Zama's current environment and the type of people they are trying to attract (hiring & growing plans).

24:47 - Agnès & Zama Outro.

25:48 - Enginears Outro.

Edited by: hunterdigital.co.uk

Hosted on Ausha. See ausha.co/privacy-policy for more information.