Emi
A calm place for people who live with migraines. Log what is happening in the moment, write it up after, and share a clean summary with your doctor.
Product Manager·London
I am a commercially minded PM with seven years building technically complex products across fraud detection, performance marketing, and property management.
When working with engineering, design and data teams, I focus on the outcome we are trying to achieve rather than the solution to build. I care about sequencing, finding the work that unblocks everything else, and keeping commercial reality in mind when product decisions get made. Feedback I have received consistently points to two things: commercial awareness and the ability to work across technical and non-technical teams.
Alongside my day job, I am building AI-powered tools to develop practical fluency in how these products are designed, scoped and shipped.

A few projects I built because the problems were real and the ideas would not go away.
A calm place for people who live with migraines. Log what is happening in the moment, write it up after, and share a clean summary with your doctor.
An AI writing tool that helps writers get unstuck at any stage of the process, without writing a single word for them. Three focused modes: plan a story, polish a scene, refine a beat.
An AI photography critique and coaching tool. Upload a photo, select the principles you want assessed, and receive structured feedback. Log in to unlock a skills profile that tracks your performance across principles over time.
An AI-powered vegetable recipe tool. Tell it what is in your fridge, describe what you are in the mood for, or hit surprise me. Get three diverse, exciting recipes every time. Log in to save favourites and personalise your experience through onboarding.
A few things I built to understand how agents and pipelines actually work in practice.
A command-line agent that researches any company and returns a structured PM briefing summary. Given a URL, it autonomously decides which pages to read, fetches the content, and writes up what it finds. Built to better understand the difference between a fixed pipeline and an agent that decides its own path at runtime.
A two-agent system that simulates a programmatic advertising negotiation between a publisher and an advertiser. Each agent operates from a defined data contract. They exchange offers, flag mismatches, and either reach a deal or walk away. When a deal is reached, the system pauses for human sign-off before anything proceeds.
“What stands out most is his ability to bridge the gap between technical and non-technical teams.”
“His commercial awareness was particularly impressive.”
“A strong instinct for where product and market meet, and his ability to navigate ambiguity while moving things forward is genuinely impressive.”