02/Approach · The product
A leadership operating system for AI-era engineering.
Four pillars, one product. Managing the augmented team is the wedge — the felt problem. Performance management instruments it and acts on it. Hiring decides who you're amplifying. Governance keeps it auditable. Each one ships something that works.
- 01
Managing the augmented team
Your best developers can suddenly do it alone — and some will wonder why they shouldn't.
Used well, AI turns ordinary developers into something close to 10x — but that's a leadership problem disguised as a productivity win. The job shifts from writing code to conducting it, and two new gaps open up: managers who were never taught to lead an AI-amplified developer (it is not the same job), and the amplified developers themselves, who quietly realise they could ship without you. Get it wrong and your seniors burn out, your juniors never build the mental models, and your strongest people start working for themselves long before they hand in notice. Get it right and the whole team levels up — and stays.
What ships A team operating model and a manager's handbook for leading AI-amplified developers: how to multiply ordinary developers with AI, how to actually manage them, and how to keep them motivated, growing, and committed to the team — not drifting off to build for themselves.
- 02
Team performance management
An honest number is worth nothing until someone acts on it.
Measuring the team truthfully is step one — and most dashboards fail even that (see the manifesto). Step two is the harder one: turning the picture into action. Which managers are quietly underwater. Who on the team has checked out and needs a real conversation this week. Which conditions — not which people — predict good engineers leaving, so you can fix them while it's still cheap. Performance management in an AI-amplified team isn't a ranking exercise; it's knowing where to put your attention before the cost shows up in someone's resignation.
What ships An honest, outcome-anchored read of team health (built on the manifesto's measurement model — DX Core 4, SPACE, DevEx in spirit, not vanity metrics), plus a manager's playbook for acting on it: who needs attention, what to do, and how to move early.
Optional — our own instruments
Where it helps, we bring measurement tooling we build ourselves. It reads conditions on the team, not verdicts on individuals: “three people here have had no career conversation in six months,” not “flight-risk: 74%.” It turns those conditions into specific manager actions — who to talk to, which team is drifting, where a strong engineer is being set up to leave. It instruments your team and sells nothing into your stack, it never scores a named person to their manager, and it's built to sit the right side of UK data-protection law. We're building it out step by step; it's an add-on to the work, never a dashboard you're sold instead of the work.
Neutral advice. Optional instruments. Never surveillance.
- 03
Hiring & interviews in the age of AI
The take-home is dead and you already know it.
Coding tests and take-homes have lost their signal — anyone can submit work that isn't theirs. AI widens the gap between strong and weak engineers rather than closing it, which makes the interview more important, not less. And the interview itself has to change: testing recall of syntax is pointless when the machine has all of it; what you're now hiring for is judgement, review skill, and the ability to reason about code they didn't write. Most teams also have no policy at all on AI in the interview — allowed, banned, or watched — so they're flying blind on the one decision that compounds hardest.
What ships A redesigned, AI-aware interview loop that tests judgement, review skill, and reasoning under realistic conditions; a recalibrated hiring bar; and a clear, defensible policy on AI use during the process.
- 04
Governing AI in the SDLC
Everyone's using it. Nobody's governing it.
The bottleneck moved from writing code to reviewing and governing it. Most organisations stall between experimentation and scale — pockets of heavy use, no shared standard, no audit trail, and a growing unease about what's shipping and who's accountable for it.
What ships A governed, auditable AI-in-the-SDLC workflow your engineers and your auditors can both stand behind.
—/Engagement
A ladder, designed so each rung earns the next.
Start small and paid. The diagnostic is the front door; programmes do the building; the advisor relationship is the annuity that opens once the value is proven.
Tier 1 · Half to one day
Engineering Leadership Diagnostic
£5,000
A scored maturity assessment across the four pillars, plus a prioritised 90-day action plan. Credited in full against a programme if you go further.
You leave knowing exactly where you stand — and what the gap is costing you.
Tier 2 · Typically 4–10 weeks
Fixed-scope programmes
£35,000–95,000
Managing the Augmented Team · Team Performance Management · Hiring & Interview Redesign · SDLC Redesign. Each ships a working artifact — a live picture of team health, a real handbook, an interview loop, a governed workflow.
Never just recommendations. Something that works, in your hands.
Tier 3 · 2–4 days / month
AI-Era Engineering Advisor
From £12,000 / month
An ongoing advisor relationship for leaders who'd rather have senior operators on call than keep rebuilding the playbook alone every month. A small number of seats, held open by invitation.
The annuity layer — opened once a programme has proven its value.
We work with a small number of teams at a time, and we don't compete on price. If the fee is the hard part of the decision, we're probably not the right fit yet — the teams that get the most from this are already losing far more than it costs to the problem it solves.
—/The boundary
What this isn’t.
The category is crowded with people who will teach your team to prompt. Useful negative space, so you know what you’re buying — and what you’re not.
- ×
Tool training. We won't teach your team to prompt Copilot, Cursor, or Claude Code — that's commoditising, and we'll refer it out.
- ×
A strategy deck. Every engagement ends with a working artifact, not a set of recommendations.
- ×
Generic “AI strategy” advice. The work is specific to how engineering teams lead, measure, and hire.
- ×
A dashboard sold instead of the work. Our instruments are an optional add-on to a real engagement — never the product you're upsold in place of it.
Start here
Not sure which rung you need?
Take the diagnostic. It tells you where you stand across all four pillars and which programme would move the needle first — before you spend a penny on the building.