Cloud Native London, July 2026

Hi folks!

Welcome to our July Cloud Native London meetup! Join us to hear from our two speakers and network with your fellow techies over pizza and drinks, or alternatively chat and following along on Youtube or LinkedIn!

6:00 Pizza and drinks
6:30 Welcome
6:45 Architecting Safe Autonomy: High-Stakes Autonomous Agents Need Deterministic Checkpoints for Safety (Hossein Kakavand, Luther Systems)
7:15 Putting Open GitOps to the test (Steve Fenton, Octopus Deploy)
7:45 Wrap up

See you there!

Cheryl (@oicheryl)

Putting Open GitOps to the test (Steve Fenton, Octopus Deploy)
The Open GitOps principles sound plausible, but the world is full of surprising realities. Join Steve Fenton and explore the GitOps research to see whether GitOps delivers on its promises, based on an analysis of over 600 perspectives from roles like DevOps engineers, developers, platform engineers, and cloud infrastructure engineers.

Steve Fenton researches the socio-technical systems behind software delivery; how organizations, people, and technology actually work together (or don’t). At Octopus Deploy, he explores GitOps, Platform Engineering, Continuous Delivery, and compliance through a lens shaped by punk history, psychology, and the kind of genre fiction that asks uncomfortable questions about systems and society.
He’s an eight-time Microsoft MVP (DevOps), a DORA Community Guide, a CD Foundation Governing Board member, and a contributor to the CNCF Platform Engineering Community Group. He’s written books on TypeScript, Octopus Deploy, and Web Operations Monitoring, and occasionally writes horror fiction.
*https://www.linkedin.com/in/stevefenton/*
https://mastodon.social/@stevefenton
*https://bsky.app/profile/stevefenton.co.uk*

Architecting Safe Autonomy: High-Stakes Autonomous Agents Need Deterministic Checkpoints for Safety (Hossein Kakavand, Luther Systems)
The more freedom we give an autonomous agent, the more it drifts into hallucinations or policy violations, yet clamping it down with prompts kills the reasoning we want. Think Brownian Motion versus a Brownian Bridge: a random walk wanders anywhere, but pin it at fixed points and it moves freely between them while always landing where it must. Deterministic checkpoints are those pins.
InsideOut applies this to infrastructure. It works with the user to shape a feature set, stack, config, Terraform and cost estimates, then deploys and manages the infrastructure in real time for monitoring, alerts and changes. Each stage must produce a structured artifact validated against fixed rules, and if a check fails only that stage repeats, catching errors before they propagate. High-stakes actions stay locked down by design: the agent can request a deploy by presenting a button, but never holds the cloud keys itself. The result is an agent that is more productive precisely because it is safely unconstrained.

Hossein Kakavand holds a Ph.D. from Stanford University. He has been with several start up in AI, ML and Distributed Systems, with IPOs on NASDAQ and LSE. He is currently a Co-Founder of Luther Systems focused on solving the Enterprise Operations problem at scale. @HosseinKavavand

Check out https://www.oicheryl.com/cloudnativelondon if you're interested in speaking or sponsoring.