70825c1a15
Week 1: The AI Multiplier (MCP + Skills) Week 2: Making AI Actually Work (CLAUDE.md) Week 3: Real Stories, Real Results Week 4: What Could You Build? (Interactive) Skeleton decks in presenterm format. PLAN.md has full series architecture, structure, speaker notes, and open decisions.
465 lines
10 KiB
Markdown
465 lines
10 KiB
Markdown
<!-- column_layout: [1, 8, 1] -->
|
|
<!-- column: 0 -->
|
|
<!-- alignment: left -->
|
|
|
|
<!-- font_size: 5 -->
|
|
<!-- column: 1 -->
|
|
<!-- jump_to_middle -->
|
|
<!-- alignment: center -->
|
|
The AI Multiplier
|
|
<!-- font_size: 2 -->
|
|
MCP + Skills in Action
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: Welcome everyone. Today we're going to talk about what happens when you stop being the middleware between AI and your systems. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- column_layout: [1, 3, 1] -->
|
|
<!-- column: 1 -->
|
|
<!-- alignment: center -->
|
|
<!-- font_size: 4 -->
|
|
The Copy-Paste Loop
|
|
===
|
|
<!-- reset_layout -->
|
|
<!-- font_size: 2 -->
|
|
<!-- alignment: left -->
|
|
<!-- new_line -->
|
|
<!-- pause -->
|
|
|
|
Sound familiar?
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
```mermaid +render +width:100%
|
|
sequenceDiagram
|
|
You->>LLM: I'm 10 minutes away from a demo, getting this error
|
|
LLM->>You: Try this...
|
|
You->>LLM: That didn't work! New error!
|
|
LLM->>You: Gather this info..
|
|
You->>LLM: Here it is...
|
|
LLM->>You: Try this...
|
|
loop
|
|
You->>LLM:😤
|
|
LLM->>You:😭
|
|
end
|
|
```
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
|
|
**You** are the middleware. You copy. You paste. You translate.
|
|
|
|
The AI is blind, deaf, and handless. You're doing all the work.
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: Walk through the frustration. Everyone in this room has lived this. -->
|
|
<!-- speaker_note: The AI is compelling you to work FOR it — gathering context, running commands, reporting back. -->
|
|
<!-- speaker_note: Complex problems need a lot of information, iteration and context. One narrow bandwidth interface doesn't cut it. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
MCP: Giving AI Hands
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
**Model Context Protocol** — open standard, created by Anthropic (2024)
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
One sentence: **MCP gives tools to AI.**
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
```mermaid +render +width:100%
|
|
sequenceDiagram
|
|
You->>LLM: I'm 10 minutes away from a demo, getting this error
|
|
LLM->>You: Let me check that..
|
|
LLM->>OCP: Get pod list
|
|
LLM->>OCP: Get pod logs
|
|
LLM->>ArgoCD: Get GitOps application
|
|
LLM->>Git: Get Git repo
|
|
LLM->>LLM: Plan fix
|
|
LLM->>You: Here is the plan. Requesting Approval
|
|
You->>LLM: Approved
|
|
LLM->>Git: Push fix to Git
|
|
LLM->>OCP: Verify change applied
|
|
LLM->>You: Applied Successfully ✅
|
|
```
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
|
|
**You** approve the plan. **AI** does the legwork.
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: Quick walk through the sequence diagram -->
|
|
<!-- speaker_note: MCP gives the LLM agency to carry out complex, multi-step plans -->
|
|
<!-- speaker_note: Collecting and collating from multiple sources -->
|
|
<!-- speaker_note: Taking actions, iterating and trying different paths when necessary -->
|
|
<!-- speaker_note: Key point: you're still in control. The AI proposes, you approve. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
What Can MCP Connect To?
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
Anything with an API. Today, in this room, we have:
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- incremental_lists: true -->
|
|
|
|
* **OpenShift / Kubernetes** — full cluster operations
|
|
<!-- new_line -->
|
|
* **Git** (Gitea, GitHub, GitLab) — repos, commits, PRs
|
|
<!-- new_line -->
|
|
* **ArgoCD** — GitOps application management
|
|
<!-- new_line -->
|
|
* **SecureTransport** — admin API for file transfer flows
|
|
<!-- new_line -->
|
|
* **Databases, monitoring, CI/CD, ticketing...**
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
<!-- font_size: 2 -->
|
|
|
|
There are 10,000+ MCP servers available. And you can build your own.
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: Don't dwell on the list. The point is: if it has an API, MCP can bridge it. -->
|
|
<!-- speaker_note: The MCP ecosystem is exploding. The standard is winning. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
Skills: Giving AI Expertise
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
MCP gives AI **tools**. Skills give AI **knowledge of how to use them well.**
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
```mermaid +render +width:100%
|
|
graph LR
|
|
subgraph WITHOUT["❌ Without Skills"]
|
|
direction TB
|
|
BP["System Prompt<br/>20,000+ tokens<br/>Everything, always"]
|
|
end
|
|
|
|
subgraph WITH["✅ With Skills"]
|
|
direction TB
|
|
AP["System Prompt<br/>~500 tokens"]
|
|
AP -->|"task matches"| L1["Load: Healthcheck<br/>+2,000 tokens"]
|
|
end
|
|
|
|
style WITHOUT fill:#ffcccc,stroke:#cc0000
|
|
style WITH fill:#ccffcc,stroke:#00cc00
|
|
style BP fill:#ff6b6b,stroke:#333,color:#fff
|
|
style AP fill:#51cf66,stroke:#333,color:#fff
|
|
style L1 fill:#4a9eff,stroke:#333,color:#fff
|
|
```
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
|
|
**Lazy-loaded expertise.** Scan descriptions → load on demand → unload when done.
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: The problem with stuffing everything into a system prompt: context pollution, cost, degraded quality. -->
|
|
<!-- speaker_note: Skills = modular SOPs. The AI loads only what it needs, when it needs it. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
A Skill is an SOP
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
<!-- column_layout: [1, 1] -->
|
|
<!-- column: 0 -->
|
|
|
|
**Without Skill** 🎲
|
|
<!-- new_line -->
|
|
|
|
> "Check if my server is secure"
|
|
<!-- new_line -->
|
|
|
|
- Runs random checks from training data
|
|
- Misses half the important ones
|
|
- Suggests changes that break your infra
|
|
|
|
<!-- pause -->
|
|
<!-- column: 1 -->
|
|
|
|
**With Skill** 📋
|
|
<!-- new_line -->
|
|
|
|
> "Check if my server is secure"
|
|
<!-- new_line -->
|
|
|
|
- Loads healthcheck SKILL.md
|
|
- Follows structured audit checklist
|
|
- Firewall → SSH → Updates → Services
|
|
- Prioritised report with specific fixes
|
|
|
|
<!-- reset_layout -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
|
|
**Your best engineer writes it once. Every agent benefits, forever.**
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
People leave. Skills persist.
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: This is the line that lands with the old guard. -->
|
|
<!-- speaker_note: Think about tribal knowledge in your team. What happens when that person goes on leave? Changes roles? Leaves the company? -->
|
|
<!-- speaker_note: Skills are institutional memory that actually works. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
Tools + Skills + Identity
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
```mermaid +render +width:100%
|
|
graph LR
|
|
T["🔧 Tools (MCP)<br/>What can I do?"] --> A["🤖 Agent"]
|
|
SP["🧠 Identity (System Prompt)<br/>Who am I?"] --> A
|
|
SK["🎯 Skills<br/>How do I do X well?"] --> A
|
|
|
|
style T fill:#ff6b6b,stroke:#333,color:#fff
|
|
style SP fill:#ffd93d,stroke:#333,color:#000
|
|
style SK fill:#4a9eff,stroke:#333,color:#fff
|
|
style A fill:#51cf66,stroke:#333,color:#fff
|
|
```
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
|
|
A workshop full of power tools, no training? Dangerous.
|
|
|
|
Power tools + SOPs + a qualified operator? **Productive.**
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: This is the summary slide before the demo. Three layers working together. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 5 -->
|
|
<!-- jump_to_middle -->
|
|
<!-- alignment: center -->
|
|
DEMO
|
|
===
|
|
<!-- font_size: 2 -->
|
|
Watch the AI. Watch me. Neither of us needs the other.
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: Three demos. Progressively impressive. -->
|
|
<!-- speaker_note: Demo 1: Cluster health check via MCP -->
|
|
<!-- speaker_note: Demo 2: Skill-driven task (healthcheck or equivalent) -->
|
|
<!-- speaker_note: Demo 3: The "figure it out" moment — AI adapts to something unexpected -->
|
|
<!-- speaker_note: IMPORTANT: While AI is working, visibly do something else. The multiplier makes itself. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
What You Didn't See
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- incremental_lists: true -->
|
|
|
|
* Every command the AI ran (and there were dozens)
|
|
<!-- new_line -->
|
|
* Every error it hit and recovered from
|
|
<!-- new_line -->
|
|
* Every tool it chose — and the ones it considered and rejected
|
|
<!-- new_line -->
|
|
* The context it carried across steps without being reminded
|
|
<!-- new_line -->
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
|
|
**That's the loop flip.**
|
|
|
|
You used to do all of that. Now you approve the plan.
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: This is the "what just happened" debrief after the demo. -->
|
|
<!-- speaker_note: The audience saw the AI work. Now name what they DIDN'T see. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
The 3x Multiplier
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
<!-- column_layout: [1, 1, 1] -->
|
|
<!-- column: 0 -->
|
|
<!-- alignment: center -->
|
|
|
|
**1x**
|
|
<!-- new_line -->
|
|
You, working.
|
|
<!-- new_line -->
|
|
The baseline.
|
|
|
|
<!-- pause -->
|
|
<!-- column: 1 -->
|
|
<!-- alignment: center -->
|
|
|
|
**2x**
|
|
<!-- new_line -->
|
|
You work + AI works.
|
|
<!-- new_line -->
|
|
Parallel output.
|
|
|
|
<!-- pause -->
|
|
<!-- column: 2 -->
|
|
<!-- alignment: center -->
|
|
|
|
**3x**
|
|
<!-- new_line -->
|
|
Parallel work + AI adds
|
|
value you didn't ask for.
|
|
<!-- new_line -->
|
|
**Compounded output.**
|
|
|
|
<!-- reset_layout -->
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
<!-- font_size: 2 -->
|
|
|
|
The person leveraging AI will outcompete the person who isn't.
|
|
|
|
This isn't a threat. It's an opportunity you're choosing to take — or not.
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: Don't make this scary. Make it exciting. The opportunity framing matters. -->
|
|
<!-- speaker_note: "Someone using AI will outperform someone who isn't" — that's the honest truth. -->
|
|
<!-- end_slide -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<!-- font_size: 4 -->
|
|
Next Week
|
|
===
|
|
<!-- font_size: 2 -->
|
|
<!-- new_line -->
|
|
<!-- pause -->
|
|
|
|
You just saw what's possible.
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
Next Friday: **how to get this for yourself.**
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
|
|
CLAUDE.md, system prompts, project context — the practical playbook
|
|
for making AI actually understand your work.
|
|
|
|
<!-- pause -->
|
|
<!-- new_line -->
|
|
<!-- alignment: center -->
|
|
<!-- font_size: 2 -->
|
|
🎯
|
|
|
|
<!-- no_footer -->
|
|
<!-- speaker_note: Tease Week 2. Keep it short. Leave them wanting more. -->
|
|
<!-- end_slide -->
|