Skip to content

Conversation

@cherusk
Copy link

@cherusk cherusk commented Dec 27, 2025

No description provided.

@github-actions
Copy link

github-actions bot commented Jan 4, 2026

This pull request has not seen any recent activity.

@github-actions github-actions bot added the stale Exempt from stale bot labeling. label Jan 4, 2026
@cherusk cherusk closed this Jan 4, 2026
@cherusk cherusk reopened this Jan 4, 2026
@HideakiImamura
Copy link
Member

Thanks for the PR. I would like to know about godon. What is the purpose and application of this software?

@cherusk
Copy link
Author

cherusk commented Jan 5, 2026

Hi @HideakiImamura

thank you for your interest and enquiry.

I had casted out a purpose document on my documentation page.

Nonetheless, I am trying to put it in abstracted notions beyond the purpose statement.

godon lives at the intersection or melding of:

  • Domain Knowledge - helps or enables infra/cloud/platform/network operations engineers mechanize performance tuning, calibration, and optimization of technology stacks
  • Democratizes advanced optimization - makes generic optimization algorithms accessible to non-experts AI for complex infrastructure configuration problems
  • Algorithm Agentic approach - behind the scenes godon bets on "agentic breeders" - domain knowledge cast in catalogs of breeders - it holds the stance that emergence from collaboration of certain diversity of multi objective metaheuristics agents and can rise to new benevolent traits in operational contexts

Unlike low-fidelity solutions like Datadog or its cousins that focus mainly on right-sizing or prediction through surrogate ML models from data, godon gathers data and acts, the high fidelity way.

You could imagine it that way:

  • godon is to dynamic, evolving configuration what Terraform (and cousins like Salt) are to static configuration.
  • godon can something akin to Performance Posture management over time

Where Infrastructure-as-Code tools handle static provisioning and declarative state, godon continuously optimizes running systems in dynamic environments - gathering data, acting, and evolving
configurations online.

Similar projects - kruize/autotune - but their focus is mostly on optimizing from within k8s clusters - at least it used to be like that at last sync point. godon with its "interactors"/"effectuators" seeks to be more generic in that regard.

Might further of interest:

  • Regarding maturity: The POC is complete with a working engine. I'm currently pushing aggressively toward easily usable MVP.

I'd be happy about stars on the project or further going partnerships! 🌟 🤝

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

stale Exempt from stale bot labeling.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants