Cloud Native London, September 2026

Hi folks!

Welcome to our September Cloud Native London meetup! Join us to hear from our three great 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 AI Inference Routing: Sovereign Control for Your AI Factory (Baptiste Assmann, HAProxy Technologies) 7:15 Telemetry for the Mind: AI Agent Observability & DORA-Driven Scoring, (Giulio Roggero, Mia-Platform) 7:45 Break 8:00 Extending Hyperledger Fabric Automation with Fablo: Lessons from Cluster Operations (Pereowei Daniel, DLM Capital Group) 8:30 Wrap up

See you there!

Cheryl (@oicheryl)

AI Inference Routing: Sovereign Control for Your AI Factory (Baptiste Assmann, HAProxy Technologies) Do you really need a SaaS AI gateway to get the most out of your inference service or AI factory? In this session and live demo, we'll take a deep dive under the hood of LLM inference mechanics to reveal how prompt dynamics, token execution phases, and GPU behaviour radically transform traffic patterns. Rather than outsourcing control to a third-party SaaS or adopting a new class of product, we'll show you how to solve these challenges with production-grade HAProxy rules: building a high-performance, fully sovereign AI gateway with battle-tested open source software on infrastructure you already control.

*Baptiste Assmann is Director of Product at HAProxy Technologies. An expert in load balancing and high-performance architectures, he plays a key role in HAProxy's evolution within the Cloud Native ecosystem. A regular speaker, he shares his expertise on the scalability and security of modern applications.*

Telemetry for the Mind: AI Agent Observability & DORA-Driven Scoring, (Giulio Roggero, Mia-Platform) We’ve moved past simple single-prompt LLM calls. Modern AI agents are multi-step state machines that invoke tools, run loops, and execute complex skills. But when an agentic session fails, how do we debug it? How do we measure its success? In this talk, we’ll treat AI agents like any other cloud-native microservice. We’ll explore how to instrument agentic sessions to track prompts, tool calls, and token usage using OpenTelemetry. We will introduce a weighted scoring framework for session outcomes inspired by DORA metrics. Finally, we will run a live, end-to-end demonstration showing an agentic session failure, how the metrics capture the bottleneck, and how we can use an "Evaluator LLM" to automatically patch and improve the agent's prompts and tools in real-time.

*25 years of experience in software engineering, serial entrepreneur with more than 10 business initiatives launched, today is co-founder and CTO at Mia-Platform, the Internal Developer Platform named by Gartner Cool Vendor for Software Engineering Technologies and in the Cloud Application Platforms Magic Quadrant. Principal focus: cloud native, platform engineering, data fabric and omnichannel experience. He likes to paint Blood Bowl miniatures, construct Lego, build and drive RC Cars and learn piano.*

Extending Hyperledger Fabric Automation with Fablo: Lessons from Cluster Operations (Pereowei Daniel, DLM Capital Group) Moving a Hyperledger Fabric network from a repeatable local setup into a cluster changes the nature of the automation problem. Network topology is only the beginning: operators must also manage identity, storage, service discovery, lifecycle state, observability, and containerised chaincode delivery. In this talk, I’ll share practical lessons from extending Fablo-based automation for real-world cluster operations. I’ll show how to preserve Fablo’s approachable, declarative developer experience while separating network bootstrap from ongoing operations, treating chaincode as an independently deployable service, and designing workflows that are easier to inspect, recover, and evolve. Attendees will leave with a concrete mental model and a set of patterns they can apply to their own…