SwingAI

The first AI-personalised golf coach

Golf has always had data. Launch monitors like TrackMan sit on ranges and courses around the world, capturing ball speed, spin rate, launch angle, club path, more numbers than most golfers know what to do with. Human coaches exist to interpret that data and turn it into something actionable. But coaching is expensive, not always available, and entirely dependent on finding the right person at the right time. SwingAI sits in the space between raw data and human expertise. The platform takes the data already being captured at the range, swing metrics, ball flight, club data, and uses AI to analyse it, interpret it, and deliver a personalised coaching report and club fitting recommendations tailored to that specific golfer's game. The player hits their shots as they normally would. The system does the rest, processing the data in real time and producing insights that would previously have required a coaching session to unlock. The tool is designed to serve the full spectrum of the game, from Sunday players looking to enjoy their round a little more, to competitive teams wanting to track and improve performance over time, to teaching professionals who want to give their students a richer, data-backed experience between lessons. The goal is not to replace human coaches but to make high quality, personalised feedback accessible to every golfer, every time they practise, regardless of whether a coach is standing next to them. SwingAI is still in development and we are building carefully: the technology, the product experience, and the right partnerships. It is the most personal project I have worked on, sitting exactly at the intersection of two things I genuinely love.

UCL MSc Management, 2025

Research submitted for my UCL MSc Management, examining the factors that influence consumer adoption of AI-powered financial advisors.

Creating AI solutions to improve workflows in Finance

3rdFloor started with a question that kept coming up in every conversation we were having around finance: why are so many of the most important processes in this industry still manual, time-consuming, and dependent on human hours that could be spent doing something more valuable? We co-founded 3rdFloor with two close friends from UCL to build AI agents that could automate financial workflows, starting with due diligence and data analysis, two areas where the gap between what exists and what is possible with AI is enormous. The thesis was straightforward: the financial industry moves enormous amounts of money on the back of processes that are still largely run on spreadsheets, PDFs, and email chains. AI agents, properly built and deployed, could compress weeks of work into hours. We built the operational infrastructure, developed relationships with early financial stakeholders, and implemented the first generation of our automation tools. 3rdFloor didn't become the company we imagined at the start, but it taught us more about building, about AI, and about the reality of the financial industry than anything else could have. The ideas we tested there still shape how I think about technology and business today.