Book a call
CorporateChaptersSaigon privateRetreatsBlueprintsBook a call
The Operating System5 Tracks · 18 ProtocolsUsed across 4 active client builds

How to build
an AI-native
team in 2026

The 18 Protocols are the operating system Edge 8 runs on every client build. Five tracks, eighteen protocols, one team that ships in days, not quarters. Read the full system below, then come learn it in person at one of our 2026 cohorts.

See the 2026 cohorts →Read the 18 Protocols
5Tracks across the operating system
18Protocols every operator learns
4Active client builds in 2026
3Cohorts across Australia and Vietnam in 2026
The Gap

Most teams use AI
few are AI-native

The difference shows up fast. Teams that "use AI" still ship at the same pace, with the same bottlenecks. AI-native teams operate on a different system. Here's where most teams get stuck.

01

Tools without a workflow

Claude open in 12 tabs. ChatGPT in another. No system, no memory, no handoff. Every day starts from scratch.

02

Engineers as the bottleneck

Every prompt, every dashboard, every integration waits on a developer queue. The team can think faster than it can ship.

The Five Tracks

One operating system
five tracks

The 18 Protocols cluster into five tracks. Read them in order or jump to whichever pressure point is closest to your team.

01-04MindsetHow AI-native teams think about software in 202605-08InfrastructureThe four-tool rails every build runs on, and the cataloguing that keeps them legible09-11PlanningHow small teams plan, then run the play in parallel12-15BuildingWhat you actually ship: UI, comms, admin, agents in the cloud16-18Hardening and continuityHow the code stays right and the work survives a handoff
The 18 Protocols

The operating system
in full

Every protocol is taught hands-on at the cohort. Reading them here builds the map. Building alongside us is how you make it real.

Track 01 Mindset, AI is the CMS

Mindset

01-04 · How AI-native teams think about software in 2026
01
CMS is dead, AI is the CMS
Content management systems were built for a world where humans typed into boxes. AI generates, structures, and updates content directly. The CMS is the model plus a folder of context.
02
The Stack: Claude, GitHub, Vercel, Supabase
Four tools, one operating model. Claude is the AI partner, GitHub is version control, Vercel is deployment, Supabase is the database. Master these four and you can build almost anything.
03
Agents: the folder is the Agent
An agent is not a magical entity. It is a folder: a role, skills, context, routines, guardrails, and a Definition of Done. Structure the folder well and the agent works.
04
The build team: humans plus AI
A team is always humans plus AI. Your build team is you, an engineer when you are stuck, and four agents that help you build: PM, Developer, QA, DevOps.
Track 02 Infrastructure, Claude GitHub Vercel Supabase

Infrastructure

05-08 · The four-tool rails every build runs on, and the cataloguing that keeps them legible
05
GitHub hygiene: check-in and check-out
Commit, push, pull, branch, merge. Checking in and out with a team, a branch per change, and Claude worktrees so several builds run at once without losing each other's work.
06
Vercel hygiene: domains and DNS
How deployments work, how custom domains attach, and the DNS basics you need so a site goes live cleanly the first time.
07
Supabase: data structure basics
Tables, rows, columns, relationships, and row-level security. Enough of the data model to read, write, and reason about a database without breaking it.
08
ABC: always be cataloguing
Folders laid out the same way every time, files whose names say what they are, and what the chat decided written down before the session ends.
Track 04 Product Management, plan in epics ship in days

Planning

09-11 · How small teams plan, then run the play in parallel
09
Prompts are dead: the plan is the prompt
Prompt engineering was typing better sentences into a box. It is over. The unit of work is a plan: an interview, a product plan in your words, a build plan for the AI, and a Definition of Done. Vague plan, average slop. Precise plan, a system.
10
Frameworks: the inside of a smart plan
What separates a smart plan from average AI slop is what you put inside it: business and academic frameworks. 100 of them in four lenses. You do not memorize them; you know they exist and point at them.
11
Multi-delegating: the build play in parallel
Plan well, build small, build all, review. Then run it several times at once: one plan per build, one worktree per build, merged in one at a time.
Track 03 Building, workflow design beats prompt engineering

Building

12-15 · What you actually ship: UI, comms, admin, agents in the cloud
12
Design systems and Claude design
Give Claude a design system and the output stays on-brand and consistent. Tokens, components, examples: what to feed in so what comes out matches.
13
Communications: email, nurture, Resend
How to send messages programmatically. Resend for transactional email, the shape of an automated nurture sequence, and the deliverability and opt-out rules that are not optional.
14
Admin interfaces and dashboards
Internal tools, a CRM, an admin UI, the Monday dashboard, so the team can read and act on the data without escalating to engineers. Synthetic data so you can see it full.
15
Agents in the cloud
A function on Vercel wakes on a schedule, reads your database, thinks with Claude, acts within its guardrails, and logs what it did. Same folder idea, different home.
Track 05 Continuity, hand off the agent not just the work

Hardening and continuity

16-18 · How the code stays right and the work survives a handoff
16
Harden the code
The development protocol, the QA plan, unit tests versus the full suite versus a human pass, and a senior code review before anything emails a customer.
17
Reflect and plan tomorrow with the PM agent
End the day by reflecting with the PM agent: what shipped, what failed QA, what slowed you down. It writes the notes and stages tomorrow.
18
Working with human tokens after you leave
How to hand off agent context, memory, and project state to the next person, so the work survives a transition without a knowledge tax.
From a current client
"We thought we were 'using AI'. Then Protocol 11, Workflow Design, landed and everything else clicked. The 18 Protocolsare the operating manual we wish we'd had two years ago."
James Murray · Work Healthy Australia
Where you learn it

Reading won't make
you AI-native
building will

A corporate retreat: we fly to your company, your team at the tables on your data, two engineers beside them, the system live before we leave. Quoted per engagement.

The corporate retreat →
Why Infinite Leverage exists · the founder story
Frequently asked

Questions
before you commit

If your question isn't here, reach out to Dave directly. Honest answers, no sales script.

What is an AI-native team?

An AI-native team is a small group that uses AI as its primary operating layer, not as a feature added on top. In the Infinite Leverage model the team is three humans (the founder, a subject matter expert at the founder's company, and an AI engineer) plus four AI agents (Product Manager, Developer, QA and DevOps). The humans hold vision, domain knowledge, and AI architecture. The agents hold the execution volume. The 18 Protocols define how the humans and the agents move together.

Do I need to know how to code?

No. The 18 Protocols are designed for founders, operators, and non-technical leaders. Engineers are in the room at every Infinite Leverage event to handle the technical depth. You bring the business and the data. They unblock you in real time.

How long does it take to master the 18 Protocols?

The protocols are walked in a single retreat, one to five days. Mastery comes from applying them on your real work. Most operators reach working fluency in 30 to 60 days after the event, especially with the White Glove tier (30 days of post-event engineering support).

Who teaches the 18 Protocols?

Dave Hajdu and the Edge 8 team. Edge 8 is the AI-native consulting team that runs on these protocols across every client build. The protocols are not theory: they are the operating manual for how the team ships in days, not quarters.

How is this different from a generic AI workshop?

Generic AI workshops teach prompts. The 18 Protocols teach a complete operating system: stack, roles, workflows, infrastructure, handoff. You leave with a working agent on your real data, not a notebook of prompt examples. Two formats (2-day build sprint in Sydney, 1-day build day in Saigon) across Australia and Vietnam in 2026.