AI agents can now carry most of the production work of building a product: the coding, the testing, the drafts. The question for a leader is what to do with that: how to organise the work, staff the team, and keep a person answerable for every result. We rebuilt our own teams to find out, on our own internal projects, and measured what happened. This page is what we learned, and how to run the change in your own organisation.
We expected the main lesson to be about speed. It turned out to be about structure, accountability, and adoption.
Most companies buy the tools, hand one to every person, and keep the process unchanged. That's AI enabled: the same handoffs, each step somewhat faster.
AI native means redesigning the process around the agents: fewer handoffs, teams organised around decisions, agents doing the production work in between.
However fast the agents produce, they don't own the result. The person who assigned the work does. Enterprise software carries consequences that reach beyond any one builder: the brand, the regulator, the customer. So every piece of work in our teams has a person's name on it, the checks before release stay as strict as ever, and the agents answer to a reviewer, never the other way around.
The math tells you what's possible. It doesn't move a single person to work differently. For a company that already exists, the real transformation is getting your people to genuinely take up new ways of working, and that happens from the ground up, through the work itself. The number to watch isn't licences bought. It's how many of your people have actually changed how they work.
In our trial runs, the speed gains were real, but the more interesting finding is where the bottleneck shifts.
Prototyping compresses dramatically: from two to three people over a week, to one person over one to two days. Where the work was clear and bounded, AI took the execution weight, and one person ran the loop end-to-end.
It is estimated that a full refactor of an existing production application is faster, when the team that owns the codebase drives the AI loop, with the right harnesses around the agents.
Reviewing AI-generated output takes more time now. Skills and harnesses are employed on the front end during code generation. Companies need to invest in specialised tools to sift through AI-generated code.
Some companies hand everyone AI and keep the org chart: the same functions, the same handoffs, everything slightly faster. The map below shows the other path. Tasks that used to require many handoffs among individuals with specialised mandates now shift to reimagined teams using agents, while the wider functions build the tools and maintain the hard lines. Read it as a process map, because that's the decision it represents.
None of these forces a smaller headcount. The same structure runs the other way: keep all your people and take on several times the work by running more of these small teams in parallel. Prototyping and experimentation become faster and cheaper. The alignment cost collapses: three people to align means three lines of agreement between them, where six functions meant fifteen. Add a person only when there is a new set of decisions to own.
The starting point is how you operate today. The destination is running as an AI native company. The five steps between them are what you implement to get across: they work in order, and each one pays off on its own.
What AI native looks like on an ordinary day
A person writes the context and requirements into the team brain once. A planning skill chops the plan into build tasks and files them in the same tracking system the company already uses. A coding agent picks them up and works in a sandbox, and the work comes back as a pull request with a person's name on the review. The whole loop runs on tools your teams already know.
When AI carries the production work, narrow specialist roles collapse into broader ones.
Decides what should exist, shows working demonstrations instead of writing documents about them, and stays responsible for the product after it ships.
Answers for the build. Used to spend the day writing code. Now spends it directing AI agents, reviewing what they produce, and designing the structure the whole system rests on.
The converging point for software engineers and data scientists. Masters the agentic development process: defines the work for the agents, sets up the harnesses they run in, and brings the technical judgment to know when their output can be trusted.
Instead of designing individual screens, builds the design standards and automated checks that let every team ship work that looks and feels like the company's products, even when AI produced it.
Beyond the responsibilities of a typical platform engineering team, the role shifts to enabling the organisation to embed best practices and guardrails for every function into the tools the company uses.
Our AI Transformation program had two goals: grow our capability in broader enterprise AI transformation work, and change how we work ourselves.
One of the simplest changes was having AI engineers own and manage their projects, with leadership supporting program management and external communications. In most cases, two people carried the work of project scoping, discovery, planning, and proof-of-concept delivery.
What emerged was rapid upskilling across the company to become fluent in agentic AI capabilities. When we removed the barriers of management defining "what needs to be done", imagination and creativity became the only limiting factor, and everyone learned how to pitch their ideas better.
Getting agents working on greenfield projects was easy; making it work in our brownfield projects is just taking off. A 2026 research paper describes a single staff engineer delivering a brownfield initiative scoped for a four-person squad in half the planned time, but that depends on a well-defined product boundary, a legible codebase, and a knowledge base that agents can reach within governance guardrails. The model becomes untenable when a decision exceeds what one engineer can hold, and it becomes a business risk when only one person understands how those decisions were made.
Any company can buy the same tools. What decides the outcome is whether your people change how they work.

What your people keep, and what you hand to the agents.
Get updates about our latest AI Transformation projects.