Home Projects Personal

Featured Projects

[contact me for code or reports where not provided]

Card image cap
Card image cap

Optimising Geospatial Data Processing
using SIMD-Accelerated Data Compression

MRes thesis on beating main-memory bandwidths for geospatial computation via cache-optimised in-memory data compression. Implemented in C++ with AVX512.
  • Deployed state-of-the-art SIMD-optimised data compression algorithms to reduce the memory footprint and improve cache efficiency during geospatial data processing.
  • Leveraged AVX512 in C++ to parallelise compression algorithms, improving performance by maintaining compressed data closer to the CPU.
  • Achieved up to 78% speedups by fusing decompression and aggregation stages of computations.
  • Utilised SLURM for managing HPC resources, enabling 8000 hours of benchmarks.
  • Applied multi-objective Pareto optimisation to narrow the search-space of codec combinations for optimal space-vs-time tradeoffs.
  • Awarded 78% (Distinction) for the thesis.

    • C++   Python   SLURM   GDAL
    Multi-Agent Car Parking Thesis

    Multi-Agent Car Parking

    Undergraduate thesis applying reinforcement learning to simulate and control groups of autonomous vehicles using PPO and Unity ML-Agents.
  • Implemented variants of Q-Learning and modifications to PPO using Tensorflow/Python.
  • Designed a flexible environment modelled as an MDP with independent agents and dynamic goals. Implemented in Unity using C#.
  • Achieved over 98% parking success with 7 cars using PPO, outperforming single-agent baselines.
  • Explored collaborative and competitive behaviours with variable communication and density.
  • Revealed novel group dynamics such as 'leaky' collaboration and competition-induced cooperation.
  • Scaled training on the university's HPC using Slurm.
  • Awarded 85% for the thesis (top 2% / 250 in cohort).

    • Python   C#   Unity   Slurm
    Card image cap

    Arbitrage Simulation

    Event-driven full-stack simulation of an arbitrage bot for betting.
  • Simulates odds changes and bet execution delays to explore slippage.
  • Distributed Python/FastAPI backend using Redis for inter-service communication.
  • Real-time TypeScript/React frontend streaming from a webhook.
  • Dockerised backend.
  • Extensively tested with 43 high-signal unit tests (28 backend, 15 frontend).

    • Python   TypeScript   Redis   React   Docker
    Card image cap

    ModulePal

    Web app enabling Warwick students to review academic modules, helping them make better decisions. It ensures legitimacy by only allowing a user to review a module they have taken (integrating with university systems), and harnesses student data to provide confidential grade-based analytics.
  • Was used semi-offically in Warwick's Computer Science department, with over 5000 reviews.
  • Harnesses Warwick student data via OAuth, personalising the experience to a user's university account.
    • Technical details:
  • Secure Spring (Java) backend, using an in-memory ORM (SQL/Hibernate) database for caching, and a NoSQL Firebase Realtime Database for persistant data storage.
  • React (TypeScript) frontend utilising hooks and other advanced features, harnessing Firebase Authentication to manage authentication states.
  • Optimised with an efficient NoSQL data schema, in-memory SQL caching layer, scalable metric computations via moving averages, and paginated queries.

    • TypeScript   Java   SQL   React   Spring   Firebase