MComp Computer Science student at the University of Bath (on track for First Class honours, recipient of Top 10 student award)
Software Engineering Intern at Microsoft (summer 2024 and 2025)
Generation Google Scholar (2023–24)
Chair of the Bath Computer Science Society (2024-25)

I’m interested in building intelligent systems, particularly agentic AI, reinforcement learning, and multi-agent systems.

• Agentic AI systems
• Reinforcement learning
• Large Language Models
• LLM orchestration and tool use

My recent work includes multi-agent reinforcement learning, skill discovery, development of agentic AI systems, and observing interactions between multiple LLM agents.

2025

• Contributed to the development of an agentic AI system using MCP servers and Semantic Kernel
• Worked in an agile engineering team delivering new AI functionality
• Presented research on multi-agent reinforcement learning to the team
• Organised intern networking events and participated in the Global Intern Hackathon

Tech: C#, Azure, Semantic Kernel, LLM orchestration, Git, Azure DevOps :contentReference[oaicite:1]{index=1}

2024

• Developed a custom Copilot solution using Azure services and Semantic Kernel
• Collaborated in an agile team and contributed to production-level code
• Presented my solution to the team and the organisation’s Vice President

Tech: C#, Azure, Azure OpenAI, Azure SQL, Git

• Led an 11-member committee
• Grew membership to become the largest subject society at the University
• Raised £10,000+ in sponsorship
• Organised a 24-hour hackathon with 200 participants and 14 sponsors

The event was nominated for University of Bath Event of the Year.

My undergraduate dissertation (awarded First-class honors) exploring graph-based skill discovery for cooperative MARL agents.

• I adapted a single-agent graph-based skill discovery algorithm to work for multiple agents working collaboratively
• I tested the algorithm against baselines in two different test environments
• I found that my algorithm outperformed the baselines in both of the environments

Through this project I learned about:

• Multi-Agent Reinforcement Learning (MARL)
• Partially Observable Markov Decision Processes (POMDP)
• Hierarchical Reinforcement Learning
• Skill Discovery and the Options Framework
• How to emprically evaluate the performance of reinforcement learning agents

Repo is private due to requirements of the course.

Multi-agent AI system where large language model agents play a social deduction game inspired by The Traitors.

Agents are given hidden roles (traitor or faithful) and must reason, form alliances, and deceive or detect deception through conversation.

Check it out here

Focus areas:

• multi-agent reasoning
• deception and trust modelling
• emergent social behaviour in LLM agents
• evaluation of LLM reasoning in social environments

Tech: Python • LLM APIs • Prompt engineering • Multi-agent simulation

Winner — Tech for Environmental Sustainability, Bath Hack 2023

Browser extension that displays product carbon emissions when shopping online.

Check out the DevPost Here

Tech

• Python
• Flask
• JavaScript
• DitchCarbon API

Winner — Best UI/UX, Women in Tech Hackathon 2024

Machine learning web app that identifies houseplants and provides care guidance.

Check out the DevPost Here

Tech

• Python
• Flask
• React
• Tailwind
• Scikit-Learn

• Generation Google Scholar (2023–24)
• Four-time Hackathon Winner
• Top 10 Computer Science student, University of Bath
• Microsoft Certified Azure Fundamentals
• Runner-up — Bournemouth Young Researchers Prize
• 3rd place — Alan Turing Cryptography Competition (1000+ teams)

Languages

Python • C# • Java • JavaScript • C • Haskell • SQL

Frameworks & Tools

React • Flask • Tailwind • Scikit-Learn • Azure • Git • GitHub

🏄 Surfing
⛷ Skiing
💃 Dance
🏃 Running
🌊 Windsurfing

💼 LinkedIn
💻 DevPost