AI and Platform Engineering? :thinking_face: Aeri...
# general
g
AI and Platform Engineering? 🤔 Aeris has been chatting with folks in the platform engineering community about this topic for a while, and they’re coming up with some pretty interesting stuff. It’s the sort of stuff you’ll want to know about before anyone else, and that you should start thinking about too. https://platformengineering.org/blog/ai-is-changing-the-future-of-platform-engineering-are-you-ready
h
Nods, got this on my list to play with for our barnraise for a custom local model for us https://docs.opencopilot.dev/welcome/getting-started - going to feed it our confluence docs too 😄
g
Super interesting @Hugo Pinheiro especially because some internal knowledge tools I tried are really not that good and I was looking for something better
p
mweh, I am a little late to the discussion, but: There is also one more perspective for Platform Engineering x Large Language Models --> implementation of platform engineering practices and mindset to improve adoption of Gen AI technology at the organization level. What we did it our company and presented on Polish Meetup of Platform Engineering Community, was usage of Platform Engineering practices to support our development teams which are building business applications powered by LLMs. Such technology introduce new challenges and high cognitive load to development teams, thus defining proper platform services which abstracts this complexity is very important:
hope is not too late ^^ let me know what do you think 😀
@Hugo Pinheiro / @Giulia Guizzardi
h
That makes perfect sense and it's a great idea 😁
p
Using e.g. Experiment Service, with interface which was defined in declarative way, developers had possibility to publish to cluster Experiment definition with: test set(list of pairs <question, expected answer>), runs(list of app config with different LLM provider, model, prompt, image version), index etc and based on it, get report as a result -> this way Development team don't need to e.g. know how we are evaluating LLM outputs and complexity is hidden another case is of course model monitoring -> the same business case which worked 3 months ago on model from provider x does not necessary will work the same now ^^