Toronto CNCF Community! 

Join us on June 19th for an exciting evening of CNCF updates, live demos, and inspiring stories from local experts using CNCF technologies.

Location

Recursion Toronto, 336 Queen St W, Toronto, ON M5V 2A2

Timeline

Please RSVP! Being on the guest list will expedite entry into the venue. Doors open at 5:00 pm, and talks begin at 6 pm. Food and refreshments are provided.

Future CFP

We're got a future event coming up in September! Interested in speaking at future events? Submit your talk through the Toronto CFP process at https://linktr.ee/cloud.native.canada.

Line Up

Yoke an Adventure into Code-First Kubernetes Resource Management

David Desmarais-Michaud - Nesto at Senior Platform Engineer

What even is GitOps or Infrastructure as Code anymore? This talk explores the current state of Kubernetes resource management, from Helm to emerging code-first tools, and how the landscape is evolving.

We'll dive into Yoke, a code-first alternative to Helm and KRO, that lets you define Kubernetes applications using general-purpose code. Together, we’ll walk through an adventure that starts with client-side resource management and ends with server-side reconciliation, aiming for the DX and reliability of a real development environment, not just templating.

Event Processing at Scale with a Kubernetes-Native Platform

Keran Yang - Staff Software Engineer at Intuit

Event-driven applications are essential for building scalable, responsive systems that handle real-time data streams efficiently. At Intuit, these architectures empower teams to react instantly to critical business events—unlocking real-time insights and driving automation across services.

In this session, we’ll explore the power of event-driven architecture using Numaflow, an open-source, Kubernetes-native platform developed at Intuit. You’ll get hands-on experience building serverless event-driven applications using Numaflow in the programming language of your choice. We’ll cover its core features, including autoscaling, seamless integration with event streams like Apache Kafka, and more.

Sandboxing Pods with runc

Jamon Camisso - Engineering Manager at Chainguard

Ever wanted to run k3s inside K8s (safely and reliably)? What about systemd, or podman or Docker? In this session, we'll explore how alternative runc implementations can be used to provide more secure and more capable pods than the default.

Defining SLOs and SLIs for ArgoCD: A Metrics-Driven Approach to Observability

Serhiy Martynenko - Senior Software Engineer at The New York Times

ArgoCD offers an extensive set of metrics that provide invaluable insights into its health and performance. However, as of now, there are no standardized guidelines or recommendations for defining Service Level Objectives (SLOs) or Service Level Indicators (SLIs) for ArgoCD. SLOs serve as a critical foundation for ensuring the reliability and quality of ArgoCD services, guiding teams to proactively address issues before they impact users.

In this talk, we will:

• Explore the wealth of metrics provided by ArgoCD components (API server, application controller, repo server, and more).

• Propose actionable SLOs and SLIs tailored to ArgoCD operations, such as deployment success rates, reconciliation time, and resource health checks.

• Demonstrate how these SLOs can help identify early signs of service degradation, enabling teams to maintain high service quality.