Writing
Notes on agentic AI, developer tooling and building things that earn trust.
Field notes from building production AI agents and from fifteen-plus years leading product and technology in regulated fintech — first-person, no fluff.
-
What a Forward Deployed Engineer Actually Does — and Why It's a Client-Facing Job
The Forward Deployed Engineer role is sold as pure engineering, but the load-bearing skill is requirements-gathering in the customer's room. What the day actually looks like, how FDE differs from solutions engineering, sales engineering and Field CTO, and how to tell a real FDE role from a support job in a fashionable title.
Read → -
What Running Product Inside a Global Bank Teaches You That a Startup Cannot
I have run product as a manager of managers inside a 50–70 person organisation at a global bank, and as the sole product and technology decision-maker at a Series A. The difference is not the amount of work — it is the kind. Notes on what scale actually changes, honestly, in both directions.
Read → -
How to Interview Someone Who Says They Direct AI Coding Agents
The interview questions that separate people who actually run AI coding agents in production from people who have used a code-completion tool and updated their CV — what to ask about the harness, review, proof and failure, with the answers each kind of candidate gives.
Read → -
What the New AI Job Titles Actually Mean — and Which Ones Are Technical
Forward Deployed Engineer, Head of AI, Chief AI Officer, AI Transformation lead — what the emerging AI job titles actually are, which are technical seats, which are rebadges of known roles, and what each one maps to.
Read → -
What Two Months of AI-Assisted Building Actually Produces
Two months of shipping daily through Claude Code and Codex multi-agent pipelines: where the speed is real, where the security isn't, and why review rigour — not model choice — is the actual differentiator.
Read → -
The Fractional CTO Test: What Founders Should Actually Check
A buyer-side guide to vetting a fractional CTO: what four days a month really buys, the evidence worth demanding, and the red flags that save you months.
Read → -
How I Run a Multi-Agent Engineering Pipeline
The concrete method behind running several AI agents in parallel on production work: worktree isolation, cross-engine review as the core gate, proof-of-work over claims, model-to-task matching, and what still goes wrong.
Read → -
Making AI Agents Reliable: Engineering Trust into Autonomous Coding Systems
How I build production AI agents that ship real software — adversarial cross-engine review, proof-of-work gates and anti-fabrication discipline that turn autonomous coding agents from demos into dependable delivery.
Read → -
Trust Is the Product: Leading Product and Technology in Regulated Fintech
Fifteen-plus years shipping regulated financial products at HSBC, Shieldpay and Zopa taught me that in fintech, trust is the product. Notes on product and technology leadership when compliance, risk and scale are non-negotiable.
Read → -
Pragmatic AI Adoption: How Engineering Teams Actually Get Value from Coding Agents
Most teams adopt AI coding tools the wrong way — chasing demos, hardcoding a single vendor, trusting output they can't verify. A pragmatic playbook for getting real, durable value from agentic AI.
Read →