ENGAGEMENT · ROLLOUT
AI Engineering Operating System
This is a consulting and implementation engagement rather than a course. I assess where your teams actually are, redesign the workflows worth changing, build the shared assets that make AI stick, and leave you with a 90-day rollout plan and the standards to run it.
Updated 19 August 2026
What you’ll learn
AI maturity assessment
An honest read of where each team is, what they already pay for, and which of the current experiments are worth keeping. Usually the first surprise is how much is already being spent.
Workflow redesign
Pick the handful of workflows where AI changes the economics, and redesign those. Everything else stays as it is, which is the part most consultants leave out.
Shared assets
CLAUDE.md and editor rules for your repositories, a prompt library that matches your material, spec templates, and MCP connections into your own tools.
Standards and governance
What is allowed, what needs review, what never goes near a model, and who decides. Written down, so adoption does not depend on individual judgement.
A 90-day rollout plan
Sequenced by department with named owners, so the work continues after I leave. This is the deliverable that separates a rollout from a workshop.
How the engagement runs
Five phases, in order. Each produces something concrete before the next begins, so you can stop at any point and still hold value.
01
Maturity assessment
Where each team actually is, what is already being paid for, and which current experiments deserve to survive. Usually the first surprise is the existing spend.
02
Workflow selection
The handful of workflows where AI changes the economics, chosen with the people who run them. Everything else is explicitly left alone.
03
Shared assets build
CLAUDE.md and editor rules for your repositories, a prompt library that matches your material, spec templates, and MCP connections into your own tools.
04
Standards and governance
What is allowed, what needs review, what never goes near a model, and who decides. Written down and agreed, not implied.
05
Rollout and handover
A 90-day plan sequenced by department with named owners, so the work continues after I leave rather than depending on me.
How it runs
Assessment first
FIXED FEE
Start with the maturity assessment alone; the readout scopes the rest before you commit to it.
Full engagement
SCOPED ON A CALL
All five phases, for companies past the experiment stage that want the operating model installed.
What you keep
- The assessment readout, including what you already spend
- Your priority workflows, redesigned with the teams that run them
- The shared asset set: CLAUDE.md, rules, prompt library, spec templates, MCP connections
- A written standards and governance document
- The 90-day rollout plan with named owners
Who it’s for
For companies past the experiment stage that need AI to become how the work is done.
- Groups rolling AI out across multiple departments or entities
- GCCs and product companies standardising engineering practice
- Leadership teams who have funded AI and cannot yet see the return
- Companies that ran training already and watched it fade
The three practices, at a glance
A · Teach
The training practice, 250+ professionals trained
- FLAGSHIP · REAL ESTATE
- SERVICE COMPANY
- COHORT
- PRIVATE GROUP
- PUBLICATION
B · Run
The adoption practice, running now
- SKILLS LIBRARY
- COWORK WORKFLOWS
- PLUGINS & MCP
- AUTOMATIONS
- ADOPTION PLAYBOOK
C · Build
The production engineering
- RAG
- AGENTIC
- AUTOMATION & MCP
- IDEA TO POC
FREQUENTLY ASKED
AI Engineering Operating System: common questions
How is this different from your training programmes?
Training changes what a team knows. This changes how the work is organised: the standards, the shared assets, the governance and the sequence of who adopts what. Training is usually part of it, but it is not the deliverable.
Does an engagement like this actually happen, or is it a package on a page?
It is the shape a real engagement grew into. Ten sessions with Shivalik Group’s IT team in Ahmedabad extended to Sales, and then to a second entity in the group. It is also how I work in my own role: as AI Lead at Sonetel I run the same operating model day to day, with an organisation-wide Skills library, MCP integrations into internal systems, scheduled automations and a written playbook that engineering, support, sales and marketing follow.
How is it priced?
Fixed fee, scoped to the engagement, shared on a short call before you commit. No hourly billing and no subcontracting. You work with me directly.
FROM THE PRACTICE
How this gets delivered, written up
WHERE THIS LEADS
What teams usually look at next
Spec-Driven Development
The loop at the centre of the operating model, taught on its own.
Claude Code Mastery
The tooling fluency the operating model assumes, for teams that do not have it yet.
AI Governance Workshop
The policy layer that has to exist before an engineering standard survives contact with legal.
Where adoption and engineering meet
THE ENGINEERING TRACK
What this looks like for an engineering team
Adoption across a company and adoption inside an engineering team are different problems. If your developers are the ones who need to change how they work, the hands-on track covers Claude Code, Cursor, agents and spec-driven development on your own repositories.
See the engineering trackTHE ROLLOUT AROUND IT
Training is the start, not the whole job
A team that learns the tools still needs standards, a prompt library, agreed workflows and a plan for the rest of the company. That is the AI Engineering Operating System engagement: assessment, workflow redesign and a 90-day rollout rather than a set of sessions.
See the rollout engagementPROOF
From ten IT sessions to a group-wide AI rollout
I ran ten AI training sessions for Shivalik Group’s IT team. The work then extended to their Sales team, and a second entity in the group is now onboarding. This is what that looked like, and why it grew.
Shivalik Group · Real estate · Ahmedabad
Read the case studyWant this for yourself or your team?
A free 30-minute discovery call is the fastest way to scope the right program. Private 1:1 or corporate, remote across India and globally, on-site in Ahmedabad.
Not ready for a call? Start with the AI adoption readiness checklist.