2026
EDB Summer Academy
Four sessions. Four critical topics. One summer to sharpen your Postgres expertise.
EDB Summer Academy is a free virtual series covering high availability, Kubernetes, AI infrastructure, and analytics at scale. Each session goes beyond the basics, with practical depth you can apply straight away.
Browse the full programme below and register for the sessions that are right for you.
Why attend?
14:00-15:00 CET | 23rd July
Stateful on Kubernetes: Running Postgres the Right Way
Kubernetes promises to simplify how you deploy and manage stateful workloads. But databases aren't stateless services, and closing that gap takes more than dropping a container image into a cluster. This session covers the core concepts DBAs and DevOps teams need to bridge the gap.
We dig into the failure scenarios teams overlook, the trade-offs between self-managed and operator-managed approaches, and what it takes to build a Kubernetes-native Postgres setup you can actually trust in production.
AI applications need a database that can handle more than rows and columns. Vectors, embeddings, semantic search, and real-time inference pipelines are now first-class database concerns. This session explores how Postgres has evolved to support AI workloads, covering pgvector and similarity search and how to structure a GenAI pipeline that keeps your data where it belongs.
We cover the architecture decisions that matter most, from data sovereignty and latency trade-offs to the operational challenges of running AI workloads alongside transactional ones. Leave with a clear picture of where Postgres fits in the modern AI stack.
11:00-12:00 CET | 4th August
Postgres as Your AI Data Platform: Architecture, Vectors, and What to Watch Out For
14:00-15:00 CET | 27th August
Stop Managing Silos: A Modern Analytics Architecture Built on Postgres
Most teams reach for a separate analytics system the moment query times start climbing but the gap between transactional and analytical workloads is narrowing. This session covers the architectural patterns that make large-scale analytics work and where ClickHouse fits as a specialist analytical engine alongside Postgres.
We address the real trade-offs between converged and specialist systems, and what you need in place to make it perform.
Most HA setups look great until something goes wrong. This session covers the core architectures, failure modes, and hard-won lessons that separate robust high availability from a false sense of security.
We dig into the scenarios teams most often overlook: replication lag during a failover, split-brain risks, fencing a demoted primary, and why automated failover can make an outage worse if not done carefully. Add in the observability and testing practices that give you real confidence in your setup, and you leave with a clear path from theory to production reality.