Julian Wiley

Training extras notes

Quick reference for training extras notes.

Context

Project: Agentic Assistants

Repository: agentic_assistants

This note is a compact reference for a local-first assistant framework with a CLI, FastAPI and WebSocket server, MCP bridge, Next.js control panel, indexing, scoped retrieval, knowledge bases, pipelines, discovery, and training workflows.

Checklist

  • Install optional extras only when needed
  • Track model runs in MLflow
  • Keep datasets local by default
  • Document quantization and adapter settings

Commands and inspection

# Start with the project README, then narrow to the subsystem docs.
rg "Training" .

# For Kubernetes-backed components, inspect the namespace before changing it.
kubectl get pods --all-namespaces

Usage note

Keep this note close to the implementation. If the command, manifest, or API changes, update the reference before writing a narrative post about the change.