ML Agent Ops Academy

A self-improving academy repository for secure MLOps and AgentOps learning, documentation, capstone development, and the Agentic AI company operating model.

Mission

Build a practical, unclassified, mission-oriented training system that helps data scientists operationalize models and agentic AI systems under secure, auditable, constrained-environment assumptions.

Core Workstreams

1. Company — Agentic AI self-improving company thesis and Customer Zero narrative.
2. Curriculum — Secure MLOps and AgentOps course design.
3. User Manual — J28 Secure MLOps and AgentOps User Manual.
4. Capstone — End-to-end MLOps + AgentOps reference implementation.
5. Agents — Governed AI-agent roles that improve the academy.
6. Evidence — RMF/ATO-style artifacts, runbooks, SOPs, and change records.

Operating Principle

Every lab attempt, error, fix, reflection, SOP, runbook, model card, data card, and change request becomes a reusable knowledge asset.

learn -> build -> document -> review -> improve -> commit -> release