NAV — ACTIVE
DEPTH 1230 M
−33.86° 151.21°E
AI Transformation — Strategy × Engineering

From AI ambition to AI in production

Built and shipped by operators — not advised from the surface;
we own the outcome

01Strategy
02AI Engineering
Megaptera novaeangliae

The humpback reads water no chart describes — and still arrives exactly where it meant to
We work the deep the same way

( The problem )

Most enterprise AI never surfaces to the P&L

Only 39% of enterprise AI pilots ever reach the P&L — the rest demo well, then drift. The reasons are predictable, and avoidable

01 / TOOL-FIRST

The tool comes before the problem

Technology-led pilots rarely move the bottom line

02 / BOLT-ON

AI bolted onto old workflows

3× more likely to pay off when built into new workflows

03 / BUILD-VS-BUY

Built what they should have bought

2× more likely to succeed when bought from specialists

( How we work )

Strategy and engineering, one senior team

The people who scope the work are the ones who build it — no handoff, no drift, from the first read to a rollout you run yourself

STRATEGY What to build Why it matters ENGINEERING How to ship it At enterprise scale MEGAPTERA LABS Strategy + build, in one room No handoff
STRATEGY
What to build · why it matters
MEGAPTERA LABS
Strategy + build, in one room
No handoff
ENGINEERING
How to ship it · at enterprise scale
01

Diagnose

Where AI actually moves the P&L — mapped fast

02

Prioritise

A sequenced roadmap; build-vs-buy decided

03

Implement

One workflow at a time, redesigned around AI

04

Scale

Embedded, measured, handed to your team

Ship in weeks, not quarters

( In production )

Outcomes that reach the P&L

A sample of what the work produces once it's live — no logos, just the numbers

End-to-end workspace

One AI workspace — sourcing to decision to tracking

1K+
Deals analysed
10+
Built-in frameworks
Productivity & velocity

Senior teams moving faster on the decisions that matter

3×
Decision velocity
60%
Manual effort removed
Revenue & margin uplift

Workflows redesigned around AI, run from inside the business

+15–25%
Revenue uplift
+5–10pp
Margin expansion
( Why now )

AI isn't digital transformation, faster — it's a different ocean: new cadence, new operating model, new kind of team — we're built for it

( Investors & operators )

We don't only advise it — we back it, and build it

We're operators and venture investors — hands-on alongside founders, and actively backing the next generation of technology being built

Venture investing Hands-on operating Deep tech
( Selected investments )
Everlab Puralink PsiQuantum
( Team )

We know the route — we've crossed it before

Senior strategists and AI builders — hands-on from first meeting to delivery

James Guo
Strategy & Transformation

James Guo

Former Bain strategist and Head of Strategy at eBay ANZ — leading enterprise strategy and operating-model redesign across financial services, healthcare, technology and consumer.

MPH, Yale
Bachelor of Engineering, University of Michigan and Shanghai Jiao Tong University

Han Xu
AI Engineering

Han Xu

Co-founder & CTO of Curious Thing AI and former Head of ML at Flamingo AI — production AI at scale for over a decade.

PhD, AI & Machine Learning, UNSW

Sam Zheng
Product & Engineering

Sam Zheng

Co-founder of Hyper Anna (Sequoia-backed; acquired by Alteryx) and Curious Thing AI; actuary, angel investor and first backer of Relevance AI.

Bachelor of Commerce (Actuarial Studies), UNSW
Fellow, Actuaries Institute

( FAQ )

Questions, answered straight

How we think about enterprise AI — and how we work.

Why do most enterprise AI pilots fail to reach the P&L?
Most enterprise AI never surfaces to the P&L — around 95% of pilots stall before they get there. The causes are predictable and avoidable: the tool is chosen before the problem is defined, AI is bolted onto existing workflows instead of redesigning them, and teams build what they should have bought.
What makes Megaptera Labs different from a typical AI consultancy?
Strategy and engineering sit in one senior team. The people who scope the work are the ones who build it — no handoff and no drift, from the first diagnostic to a production rollout your own team can run. We own the outcome, not just the advice.
How does Megaptera Labs work with clients?
Four steps. Diagnose where AI actually moves the P&L; Prioritise into a sequenced roadmap with build-vs-buy decided; Implement one workflow at a time, redesigned around AI; and Scale — embedded, measured, and handed to your team.
Should we build or buy our enterprise AI?
It depends on the workflow, and deciding it deliberately is part of the work. Specialist partners succeed roughly twice as often as in-house builds, and redesigning the workflow rather than bolting AI on makes EBIT impact about 3× more likely — so we settle build-vs-buy early, per use case, instead of defaulting to a custom build.
Where is Megaptera Labs based, and who do you work with?
We're based in Australia and work with enterprises and founders across financial services, healthcare, technology and consumer. Beyond advising, we back and build — operators and venture investors active in deep tech.
How do we get started?
Book a short discovery call and we'll give you a straight read on where AI lands — and where it doesn't. Get in touch below.

Let's get you to production

We'll give you a straight read on where AI lands — and where it doesn't

Book a short discovery call to get started