1 / -
navigate
LIVE INTRO · FOR PARTNERS · NO CODE REQUIRED

Become an AI-first operator

Ship real customer work without writing a line of code. A live run-through of the power tools.

Scroll or press ↓ to begin
WHAT PARTNERS ARE TELLING US

Three things we keep hearing.

Different firms, different regions, same three patterns in every partner conversation this year.

🎯
Talent is stretched thin
"We are short on skilled consultants, and the ones we have spend a quarter of their week on the boring work."
📈
Demand is ahead of staffing
"Customers are not asking if agents fit. They are asking how many they can buy. We are selling agents faster than we can staff them."
Long projects carry risk
"Modernisation projects are long, expensive, and the outcomes are not guaranteed. Every long project is a bet."

If you recognise one of these, you are not alone. If you recognise all three, this session is for you.

THE HOOK

Reclaim 25% of your team's time

Three numbers every partner leader should know. One of them is already moving.

60–80h
Documentation overhead per engagement (Forrester)
15%
Consultant time on unbilled admin (McKinsey)
25–40%
Overhead reduction possible with AI (GitHub Octoverse)
Public proof point
Accenture's published numbers. Pull-request cycle time dropped from 9.6 days to 2.4 days. 84% more builds passing CI. The productivity case for AI-first delivery is no longer theoretical, and it is public. GitHub research →
PROOF + IMPLICATION

Turn your delivery model into a competitive advantage

A Microsoft partner already shipped this. Earlier this year. The question is what your practice does with the same play.

Anonymized partner case · this year
A full Azure data platform. About one week.
  • ~1 week zero to production
  • 1 delivery lead + fleet of specialist agents
  • Build AND operate. Same fleet runs it after go-live
  • Recurring on operate outsizes the build fee in year one
What this means for your practice
Three plays open up.
  • Compress the build fee. Days, not months. Win on time-to-value, not body count.
  • Build IP that compounds. Agents, skills, prompts and templates you ship for one customer become reusable assets across the next ten. Build once. Ship many times.
  • Earn on outcomes and on operate. Fleet stays on after go-live. IP becomes recurring revenue. Outcome-based offers and pricing land where the platform delivers measurable KPIs.
Different delivery model. Different business model. The first one enables the second. Full customer story →
WHAT IS POSSIBLE NOW

What an AI-first operator does

Same tool. Plain English. From a quick brief to a working prototype, none of it requires code.

Easier to start More advanced
📥
Research → briefs
A messy transcript or a pile of research becomes a customer brief, grounded and ready to send.
🎤
Next-level presentations
Notes in, a deck that presents in the browser out. Send it as one link. No template wrestling.
🔁
Automated content flow
Recaps, follow-up emails, status updates. The repetitive writing starts to run itself.
🧪
Prototypes for demos
Spin up a working, clickable demo for a customer meeting. Branded and real, not a screenshot on a slide.

Pick the one closest to your week. That is where you start tomorrow.

LIVE

Let me show you one.

Messy notes in. A briefing ready to send, out.

LIVE DEMO

Messy meeting notes → a briefing ready to send

Not one giant prompt. A back-and-forth: I ask, it reasons and calls tools, it comes back. I open the real result at the build.

No template. No copy-paste from five tabs. A single web page you can send by link or present in a browser. The research is grounded in public sources, not guessed.

WHAT JUST HAPPENED

A short conversation. Three moves you can reuse.

Forget the briefing for a second. This is the pattern underneath, and it holds whatever the output is.

↩️
Restate
It says what it understood before it works. You catch a misread in seconds, not after twenty minutes.
🔎
Research
It grounds the content in a real public source instead of guessing. The claims hold up.
🧱
Build
It produces a real file you can send. Not a chat answer you copy-paste somewhere else.

Same rhythm every time. Different input, different output. That is the whole thing.

THE TOOLKIT

Four pieces. That is the whole kit.

Under the hood of everything I just showed you. None of it needs code.

⌨️
Copilot CLI
One place to type plain English. Unlike a chat box, it reads and writes files on your machine.
🧩
Skills
Reusable instructions. Teach it your way of working once, reuse it forever. The briefing was a skill.
🔌
MCP connectors
Links to your tools. Outlook, Teams, your data, your systems. It works where you already work.
📂
Files as context
Point it at a folder. Your notes, transcripts and brand become the input it works from.

You describe the outcome. It does the work.

TAKE IT HOME

Everything I showed you, yours to keep.

A repo, a course, and a one-click setup for locked-down laptops.

📦
Public repo
Skills, examples, and the briefing exercise you just saw.
github.com/JW-Sthlm/
ai-operator-intro
🎓
Course site
Four short lessons, written for partner scenarios. Not for developers.
jw-sthlm.github.io/
ai-operator-intro/course
☁️
One-click Codespaces
Laptop locked down? Run all of it in the browser. Nothing to install.
included in the repo
Your first move tomorrow
Pick one task you do every week and do not enjoy. The recap, the status update, the meeting brief. Point Copilot CLI at that one thing. Not a project. One task.

Still not a developer.

I just stopped doing by hand what a tool can do in a minute. Point it at one task tomorrow.

Johan Wallquist
Johan Wallquist
Partner Solution Architect · Microsoft Data & AI

Decks, exercises, links: partner page. One message away for the rest.

Waiting for presentation...
- / -
Connecting to presentation…
LIVE DEMO · MEETING NOTES → BRIEFING
Copilot CLI · live · a conversation
PS C:\halcyon> copilot --allow-all-tools
GitHub Copilot CLI 1.0.57 · ready
type your ask in plain English
~/halcyon-briefing claude-opus-4.8
Find my latest meeting notes in this folder and read them. No slides yet. One sentence: what does Helena need the LT to believe by Thursday?
do not type the path. let it go find the file…
readmeeting-notes.md · found in ~/halcyon-briefing
restate4 themes · 2 decisions · the real split
Not “we need an AI strategy.” The shift: delivery model and business model move together.
Go to Microsoft's Work Trend Index, find the latest report, and pull the Frontier vs Stalled gap. Then a couple of public case studies for a firm this size.
opinions are cheap. go find the receipts…
fetchworklab/work-trend-index · WTI 2026
web_searchFrontier firm case studies · advisory
Every number now traces back to a source I can name out loud.
Five in the LT: CEO, CFO, Delivery, People, Growth. What is the through-line for all five, and which disagreement do I have to resolve in the deck?
five people, five wants. find the through-line…
map5 stakeholders · Helena · David · Marcus · Sara · Tom
They split on “add AI” vs “rebuild.” The deck has to pick a side.
Propose a 6-slide outline, one argument per slide, no bullet lists. CFO leaves with a number, CEO with a story. Then tell me the weakest slide.
draft it, then attack my own draft…
outline6 slides · one argument each
critiqueweakest slide · the CFO number is buried
It found its own soft spot before I did.
Build it. HTML, Frontier style, dark, amber accent. Output briefing.html with presenter view. Open it.
writebriefing.html · Frontier style · amber
buildpresenter view ready · opening in browser…