About
I’m a full-stack software engineer at Twilio who loves all things code. I have a deep passion for building performant products that delight customers and love learning new things. Some of the languages/frameworks that I’m using right now are: Java, Python, TypeScript and React.
I recently graduated with an MEng in Computer Science with First Class Honours (Dean’s List) from University College London. In my dissertation, I researched techniques for enabling populations of AI agents to learn to cooperate using multi-agent reinforcement learning.
Outside of work, I love reading, playing board games and watching funny cat videos. I’m also currently teaching myself to play the piano.
Experience
Software Engineer @ Twilio (Aug 2022 - Present)
Developing full-stack AI features for Twilio Flex, a programmable contact center.
Software Engineer Intern @ Twilio (Jun - Aug 2021)
Built accessibility into the Twilio Flex UI by adding new functionality and refactoring the entire codebase to ensure that Flex became fully keyboard-navigable, worked well with a screen reader and handled errors and notifications in an accessible way. This work also led to the creation of accessibility guidelines for Flex UI.
After I finished my internship project, I also refactored the Flex Diagnostics Plugin and updated its UI to meet Twilio Paste standards.
Software Engineer Intern @ Publicis Sapient (Jun - Aug 2020)
Developed a full-stack progressive web app that directs customers along the most efficient route through a Tesco store, based on the contents of their shopping list.
Featured Projects
Investigating the Impact of Direct Punishment on the Emergence of Cooperation in Multi-Agent Reinforcement Learning Systems
Paper •
As the use of AI becomes more pervasive throughout society, the need for socially intelligent agents that are able to navigate complex cooperative dilemmas is becoming increasingly evident. This research studies the impact of social mechanisms such as punishment, partner selection and reputation on the evolution of cooperative behaviours within multi-agent reinforcment learning systems.
Of Mice and Messages
Worked in conjunction with Microsoft Enable to design and develop a fun and interactive eye tracking game, designed for individuals with severely disabling conditions such as ALS and Spinal Cord injuries. We created the game using Unity and C# and developed a custom Unity plugin for Tobii eye trackers before releasing the game onto the Microsoft Store.
KORTED - Kind Of Real Time Event Detection
Developed a novel real-time event detection algorithm that uses social media data. KORTED correctly identified 88% of events within less than 5 minutes from the peak of the ground truth event.
FHIR Analytics Dashboard
This project was one of the winners of the GOSH FHIRworks2020 hackathon. It’s a realtime web dashboard that pulls hospital records from a FHIR server and produces analytical visualisations using React and Recharts.