The conversation started from a familiar enterprise problem: teams building microservices in isolation end up duplicating work. You talk to one team and they have a great microservice design. You talk to another team in the same organization and they have built the same thing. There is a lot of overlap, and nobody has a clear picture of what already exists.
This gets worse the bigger you get, especially when you empower teams to choose their own tools. That freedom is good for autonomy but leads to tool sprawl – fifteen tools for analytics, fifteen for CRM, fifteen for monitoring. The same pattern shows up in vulnerability scanning, where someone built a tool just to aggregate results across all the other scanning tools to avoid surfacing the same finding multiple times.
Automatic service discovery sounds like the answer, but I have never seen it work perfectly. Google understood this well – you need a combination of automation and human curation. Clicks and surfacing behavior help, but there is always a human aspect that drives both sides. It is the same challenge as search engines: you need some top-down creation alongside the bottom-up discovery.
Watch on YouTube – available on the jedi4ever channel
This summary was generated using AI based on the auto-generated transcript.