
An agent will write you a CDK stack, a Terraform module, or a stack of IAM policies in seconds.
Whether any of it works is a separate question, and the usual way to find out is to deploy to a real AWS account and watch what breaks.
In an agentic workflow, that means giving AI access to a public cloud account, racking up costs on the AWS bill, and waiting for provisioning to complete every time you push new code to the environment.

Silvio and Carole introduce lstk, LocalStack's next-generation CLI built from scratch in Go. They explain why the team rebuilt the CLI, walk through its key features — zero-config startup, seamless browser-based authentication, and a rich interactive TUI that surfaces real-time progress and actionable errors — and demo how lstk gets you from install to a running emulator in seconds. They also dive into the architecture behind lstk and how it is designed to support multiple emulators, flexible runtimes, and deep integrations with CI pipelines and IDEs.

Cloud-first development kills your inner dev loop. There's a better way. TypeDB on AWS is powerful — Lambda functions querying complex relationship hierarchies, recursive schema functions resolving transitive memberships, all without application logic. But iterating on schemas and queries in the cloud is slow and expensive. Harsh Mishra introduces the TypeDB extension for LocalStack: run a fully functional TypeDB server inside your local AWS environment and connect your app exactly as you would in production. Faster iteration, zero unnecessary spend.

Are Property Graphs living up to the hype? Maybe the model itself is the problem. We made the move from relational databases to graph databases to escape "Join Pain" and model the real world more naturally — but for many engineering teams, that promise has curdled into something worse: the Spaghetti Graph. Complex queries. Ugly workarounds for multi-party relationships. Fragile schemas that shatter with every iteration and become a nightmare to maintain. The good news? The problem isn't your data. In this talk, Joshua Send breaks down why standard Labeled Property Graphs (LPGs) fall short when applied to complex domains — and introduces TypeDB, a strongly-typed database that brings together the connectivity of a graph with the integrity of a relational model. You'll come away understanding why LPGs struggle at scale and complexity, what "Spaghetti Graphs" are and how teams fall into the trap, how TypeDB's type system enforces data integrity without sacrificing flexibility, and when a strongly-typed graph database is the right tool for the job. Whether you're deep in a graph migration, evaluating database architectures, or just tired of schema chaos — this one's for you.