About
π Iβm currently a co-founder and CTO at Gradient Labs; weβre redefining the future of customer support with LLM agents.
π¦ I previously spent five+ years building the machine learning discipline, platform, and use cases at Monzo Bank in London. I held various titles along the way, from Data Scientist through to Director of Machine Learning and Staff Machine Learning Engineer; I reviewed this journey in 2020, 2021, 2022 and also wrote about Monzoβs machine learning platform.
βοΈ Before joining Monzo, I was a Senior Data Scientist at Skyscanner, where I built recommender and ranking systems to improve travel information in the app.
π« Before Skyscanner, I was a Senior Research Associate in the Computer Lab at the University of Cambridge, working on healthcare mobile apps that use smartphone sensors. I spun out this research into a startup that was part of Accelerate Cambridge in the Judge Business School.
π I did an MSci in Computer Science, PhD on recommmender systems, and first postdoctoral research position on urban data science in the Department of Computer Science at University College London, where I am still an Honorary Research Associate. While at UCL, I also spent time as a visiting researcher at Telefonica Research, Barcelona and worked as a Data Science consultant.
π‘ My work has always focused on systems that use machine learning - this has taken me from recommender systems to urban computing and travel information systems, digital health monitoring, smartphone sensors, banking, and open source machine learning tools. You can read more about my work and research in the Press & Speaking and Research sections.
π¨π½βπ» Open source projects
- An open source model registry: modelstore
- Audit AI APIs: ai-auditor-cron and ai-auditor-web - Award for Greatest Potential in the Stanford AI Audit Challenge
- A CLI tool for creating & deploying containers: kettle
- A cron job to destroy unpopular tweets: tweet-destroyer
- Send a telegram message when a classifier detects sound: BeepNet
- Collect & classify accelerometer data: pydata-2016