Blog
Notes on reliable event infrastructure.
Product updates, architecture notes, and workflow patterns for teams building on Hooksbase.
Featured
AI agent vs AI workflow: what is the difference?
An AI workflow has a fixed path; the model fills in the blanks. An AI agent decides the path itself. Here is the difference, when each fits, and the hybrid that survives production.
AI workflow automation: a practical guide
AI workflow automation combines LLMs with deterministic steps to handle repetitive work. Here is the practical playbook: pick the right shape, decide where the LLM goes, set up reliable triggers, ship, and recover when it breaks.
Event-driven architecture for AI agents (a practical primer)
Event-driven architecture lets components communicate by producing and consuming events instead of calling each other directly. AI agents push every system that uses them toward EDA — here is what it is, the patterns that work in production, and the failure modes you have to design for.
Event sourcing in plain English (for engineers shipping agents)
Event sourcing stores every change as an immutable event and rebuilds state by replaying them — giving you free time travel, complete audit trails, and the ability to replay against new logic. Here is the pattern in plain English, and how it relates to AI agent replay.
What is a message queue? (and when webhooks are a better fit)
A message queue is a durable buffer between producers and consumers — it decouples speed, survives outages, and supports fan-out. Here is how queues work, when they fit, and when webhooks beat them.