FIRESIDE CHAT: Can ‘MLSecOps’ do for MLOps, what DevSecOps is doing for DevOps?

By Byron V. Acohido

Massively interconnected digital services could someday soon save the planet and improve the lives of one and all.

But first, enterprises and small businesses, alike, must come to grips with software vulnerabilities that are cropping up – and being exploited – at a blistering pace.

Innovative vulnerability management solutions are taking shape to meet this challenge. One the newest and most promising spins out of the emerging discipline of machine learning operations, or MLOps.

One supplier in the thick of this development is a Seattle-based start-up, Protect AI.

Guest expert: D Dehghanpisheh, co-founder and CRO, Protect AI

I had the chance recently to visit with Daryan Dehghanpisheh, whose professional experience prior to co-founding Protect AI includes four years as the Global Leader of AI/ML Solution Architects at Amazon Web Services.

Protect AI launched in December 2022 with a  $13.5 million seed round stake, co-led by Acrew Capital and boldstart ventures, on the basis of  developing advanced tools to protect AI systems and machine learning models.

We discussed how the fledgling field of MLSecOps parallels the arrival and maturation of DevSecOps. “DevSecOps is putting security at the heart of everything you do from a DevOps perspective,” Dehghanpisheh told me. “We want to do the same thing with MLOps . . . treat security as an integral part of development, not just as an afterthought”

For a full drill down on how Protect AI hopes to mainstream MLSecOps – and how that could accelerate the arrival of massively interconnected digital systems — please give the accompanying podcast a listen.


Pulitzer Prize-winning business journalist Byron V. Acohido is dedicated to fostering public awareness about how to make the Internet as private and secure as it ought to be.

(LW provides consulting services to the vendors we cover.)

January 25th, 2023