Disaster Recovery in the Serverless Realm

High availability is table stakes in modern serverless systems, but true disaster recovery requires intentional design choices driven by business requirements. This session presents a pragmatic approach to disaster recovery for AWS serverless architectures, focusing on measurable outcomes such as RTO and RPO rather than defaulting to multi-Region complexity.

We’ll examine resilience patterns within a single Region, including retry strategies, asynchronous workflows, idempotency, and failure isolation. From there, we’ll analyze multi-Region disaster recovery models for serverless workloads, comparing active/passive and active/active architectures and their implications on data consistency, traffic routing, and observability.

This talk equips architects and senior engineers with a decision-making framework to design disaster recovery strategies that are cost-aware, operationally realistic, and aligned with business risk.

Pay Less For Serverless: Practical Tips

Serverless architecture provides scalability, flexibility, and operational simplicity, but costs can quickly spiral out of control in large-scale environments. This session will dive into practical cost optimization strategies for AWS Lambda and other serverless services, covering key areas such as execution time tuning, provisioning strategies, architecture best practices, and AWS-native cost-saving mechanisms. Attendees will gain actionable insights into reducing cloud expenses while maintaining performance, including optimizing memory allocation, leveraging Step Functions, reducing logging costs, and utilizing Compute Savings Plans.

The Hard Truth About Platform Engineering Adoption

Many organizations talk about Platform Engineering, but few are prepared for what it actually takes to move from a traditional DevOps team to a real platform team.

In this talk, I’ll share real lessons learned from transforming a small DevOps team supporting over 200 engineers into a Platform Engineering team. We’ll walk through what worked, what failed, and where theory broke down when faced with legacy services, urgent requests, and rapid company growth.

We’ll cover how self service platforms reduce bottlenecks but fail without standardization, why Golden Paths are critical but painful to adopt, and why enforcement is often the missing piece no one likes to talk about. This session focuses on practical decisions, trade offs, and mistakes, not idealized architectures.

Attendees will leave with concrete guidance on where to start, what to avoid, and how to think about Platform Engineering as a mindset rather than a single tool or portal.

The 10 Commandments of Working in Production

What scares engineers the most? Not spiders. Not clowns. Production incidents.

A single misstep can take down critical systems, impact customers, and put your name on a P0 incident. While outages are often seen as inevitable, many of them are preventable with the right mindset and practices.

In this talk, I’ll share the “10 Commandments of Working in Production” — a practical, experience-based framework for reducing risk, improving reliability, and building confidence when operating live systems.

Through real-world examples (including painful mistakes), we’ll explore how small decisions in day-to-day work can have massive production impact, and how to avoid them.

Key Takeaways:

How to approach production with a safety-first mindset
Why rollback plans are non-negotiable (and how to build them)
The hidden risks of “harmless” changes
Best practices for deployments across environments (Dev → Staging → Prod)
Why timing, collaboration, and version control matter more than you think
The dangers of manual changes, root access, and unchecked AI-generated code
How to turn production incidents into long-term engineering growth

Audience:
DevOps Engineers
Backend Developers
SREs
Engineers working with cloud and production systems (any level, especially mid to senior)

Session Level:
Intermediate

Why This Talk:

Many engineers learn production the hard way — by breaking it.

This talk compresses years of real-world experience into actionable principles that can help engineers avoid common pitfalls, reduce outages, and operate with more confidence in high-stakes environments.