AI-101 — Systems by Layer
For network engineers, DevOps, and technically serious implementers
⚠️ SYSTEMIC OVERVIEW — READ FIRST
This is a systemic mapping of the AI ecosystem, structured like an OSI/RFC layer model.
Every serious engineer on this team will eventually go neck-deep into some AI tool or vendor ecosystem for code, chat, or making something. This exists so you know exactly what kind of shit you are walking into — layer by layer: what you own, what you rent from vendor, what you borrow from OSS toy, and what disappears the moment vendors want to screw you over for their profit.
Without looking at the lego blocks, you are not making technical decisions. You are an engineer trusting marketing and Facebook KOLs.
Every serious engineer on this team will eventually go neck-deep into some AI tool or vendor ecosystem for code, chat, or making something. This exists so you know exactly what kind of shit you are walking into — layer by layer: what you own, what you rent from vendor, what you borrow from OSS toy, and what disappears the moment vendors want to screw you over for their profit.
Without looking at the lego blocks, you are not making technical decisions. You are an engineer trusting marketing and Facebook KOLs.
WE WILL NOT BE TALKING ABOUT THIS, GO ASK YOUR CHATBOT :
- How training models actually work, weights, decoders, or transformer math.
- Product hype, Google toys vs ChatGPT, making slideshows, or TTS/STT bullcrap.
- How to install OpenClaw, use Claude Code, or mess around with Antigravity.
Model ≠ Runtime
Runtime ≠ Surface
Tool Call ≠ MCP
Subscription ≠ API
Unofficial ≠ Foundation
Control layers > brand names
Part 5 — Surface
"Where you talk. Not where it thinks."
Part 2 — Runtime Layer
The Control Plane
framing
orchestration
state
tools
MCP / ACP
approvals
multi-agent
"The model is the data plane. The runtime is the control plane."
6. Control
"Is this allowed?"
Auth · TOS · Official path vs opportunistic · Sub ≠ API
5 mod
7. Economics
"Can you pay?"
API vs Sub vs Local · Cost routing · Flat-rate ≠ portable
6 mod
4. Context / RAG
"Not memory. Fuel."
- Retrieved, ranked, budgeted, injected — never stored
- RAG ≠ memory (stale source = stale output)
- Wrong chunk = confidently wrong answer
1. Inference
Stateless Processor
"Receives. Processes. Returns. Forgets."
- Stateless processor — no memory between calls
- 1M context ≠ 1M reliable recall (degrades at high fill)
- Runtime manages state, not the model
3. Ext Services
OCR
Search
Crawl
Code
Embed
"Consumed, not owned."
- Consumed by runtime — not owned
- Swappable via MCP, HTTP, or SDK
- OCR (capability) ≠ Router ≠ Protocol