Files
mik-tf c769942374 init
2025-11-25 15:27:56 -05:00

88 lines
2.9 KiB
Markdown
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

---
sidebar_position: 3
---
# Agent Use Cases
This page adapts the **Agent Use Cases** section from the Mycelium site into documentation form.
## What the Framework Enables
> *“The framework gives you full control over where agents live, how they connect, and what data they use.”*
Agents are intended to be:
- **Locationaware** you decide which nodes or VDCs they inhabit.
- **Networkaware** connectivity and topology are visible and controllable.
- **Dataaware** data stays where it should, with explicit movement.
## Example Use Cases
### Run Agents on Your Own Hardware
Deploy agents directly on:
- Laptops and workstations.
- Homelabs and edge boxes.
- VDCs and clusters you control.
You are not tied to any single cloud or vendor; agents execute where it makes the most sense for latency, privacy, or cost.
### Connect Agents Over the Mycelium Network
Use **Mycelium Network** as the secure fabric between agents.
- Private overlay addressing across homes, offices, and datacenters.
- Encrypted paths between nodes in different countries or environments.
- One consistent address space spanning local and remote infrastructure.
### Keep Data and Memory Private by Default
Agents are designed to:
- Use local datasets, tools, prompts, and embeddings.
- Avoid sending sensitive context to external AI APIs by default.
- Respect policies for which data may be shared and where.
This is especially important for regulated sectors and internal systems.
### Build Workflows Across Cloud + Edge
Orchestrate multinode workflows that span:
- Edge clusters near data sources.
- Central VDCs for heavier compute or aggregation.
- Personal devices for interaction and control.
Examples:
- Realtime processing at the edge, with summarized results aggregated in a VDC.
- Agents coordinating tasks between office locations and remote workers.
- Data pipelines that never leave your chosen infrastructure.
### Operate in Regulated Contexts
Run agents where strict requirements apply:
- Healthcare, finance, public sector, and other regulated domains.
- Environments with **data residency** constraints.
- Workloads requiring verified identity and controlled connectivity.
The combination of **local execution**, **sovereign infrastructure**, and **encrypted mesh networking** is intended to make regulatory compliance more tractable.
### Blend Local and Remote Intelligence
Not all tasks need to run on the same node:
- Lightweight agents might run locally for responsiveness and interactivity.
- Heavier workloads can be scheduled to trusted remote nodes (e.g. GPU clusters).
- Data movement is explicit and policydriven.
This supports hybrid strategies where you balance privacy, performance, and cost.
---
## Looking Ahead
These use cases are **illustrative** and may expand as the Agent Framework matures. For how to prepare your infrastructure today, see **[Getting Ready for Agents](/ai-agent-framework/getting-ready)**.