Board Director · NED · AI Advisor · London
Twenty years building production AI across six startups, one 11,000-person energy company, and a Facebook team that shipped $2.5B in incremental revenue. Three co-founder roles. Four exits. I now sit on boards and advise the people making the calls.
Input: your hardest technical question · Output: a decision the board can act on
Built, led & shipped at
I take a small number of engagements at a time, so the work is always substantive. Pick the port that matches your need — the work shapes itself accordingly.
Port A — Founders
For Series A–C founders building AI-native products. A board-level operator who's already done the things you're about to do — and can say no to the things that look right on paper but kill velocity.
Port B — VCs
Drop me onto the boards of your AI portfolio companies as a Non-Executive Director — the grown-up technical voice your founders need. Plus pre-investment diligence covering the things founders won't tell you and the questions junior associates don't know to ask.
Port C — PE firms
For mid-market and growth PE. I take board seats on portfolio companies and help turn the AI conversation from cost-cut theatre into actual EBITDA. I've run engineering inside a PE-backed £1B+ business through 10× scale — I know where the value sits.
Concrete shapes of work, all at board or advisory level. The right configuration depends on your stage and what your board is missing — we'll figure it out on the call.
Out of scope. I don't do fractional CTO, interim engineering leadership, or hands-on contract work. My day job is Kaizan — the rest of my time is reserved for board, NED, and advisory engagements where the leverage is highest.
This isn't advice from the sidelines. I've implemented these frameworks at Kaizan and helped other startups do the same — companies where AI is the operating model, not a feature bolted on.
Most companies add AI as an afterthought. The ones that win build it as the operating model. At Kaizan we wired AI into every business function — engineering, sales, customer success, ops — so every team member runs at 5–10× the leverage of a conventional hire. I'll help you design that system from scratch or retrofit it into what you already have.
The org chart most startups use was designed for a pre-LLM world. The right hiring order, role definitions and reporting lines for an AI-first company look different. I've built and scaled AI engineering teams from 2 to 70+, and led 700+ engineers across OVO Group's business units. I'll show you what works — and what looks good on paper but kills velocity.
I've run high-output engineering across eleven countries on four continents simultaneously. Remote-first isn't about videoconferencing — it's about async culture, AI-powered coordination, and eliminating the meeting tax. I've refined this playbook across six startups and one 11,000-person company scaling through 10×.
The advice comes from doing the work — across six startups, one global ad platform, and a PE-backed energy giant.
AI-native platform for enterprise client service. Kaizan helps agencies and managed-service businesses protect and grow revenue from existing client relationships — turning the calls, emails, and meetings already flowing across accounts into stakeholder intelligence, health scoring, and autonomous agents that act on the signal. Built the product, the AI architecture, and the engineering organisation from zero. Multi-agent on Anthropic, OpenAI, and open-source LLMs. £3M+ raised, 20-person team.
UK's largest energy provider. $250M+ raised. Led engineering across three business units during 10× company growth (1,100 → 11,000 employees) and expansion across UK, Europe, South Africa and Australia. CTO at Kantan, OVO's first incubator startup, which I co-built with the founding CEO. Participated in six+ acquisition due diligences, including SSE.
Led a 46-person team across two continents. Shipped Brand Awareness Objective, Reach Objective, Brand Polling, and Creative Hub — generating $2.5B in incremental revenue. Drove Facebook Audience Network growth via new mobile ad formats, lifting profit margin +35%. Scaled team from 12 to 46.
Personalised AdTech platform processing real-time behavioural data for 85% of UK internet users. $7M+ raised. Grew from 2 to 70+ employees across UK, Singapore and Australia. Won 6 industry awards.
Content monetisation platform. Built Skimwords as the third engineering hire.
C2C marketplace co-founded and exited. Engineering at Gupshup (now a unicorn) and Rediff.com. NLP & information-extraction research at the Center for Indian Language Technology, IIT Bombay.
Content Delivery Optimisation Using Exposure Memory Prediction · USPTO 20190188740 · 2019
Electronic Displays with Customised Content · USPTO 20180181412 · 2018
Building AI-powered products since 2003. LLM products on Anthropic and OpenAI since 2020. Areas I'm hands-on with today:
Selected portfolio. I back founders building category-defining AI products — and sometimes take a board seat alongside the cheque.
A handful of things I've written for fellow operators.
Let's talk
Whether it's a board seat, NED introduction, or a due diligence sprint — it always starts with thirty minutes on the calendar.