Edward O’Dwyer

edward (dot) o-dwyer [at] warwick.ac.uk · in/edward-odwyer · github.com/edwardodp

Technical Skills

Languages: Java, Python, C/C++, SQL (PostgreSQL & SQLite), Shell script, Haskell

Dev. Tools/Frameworks: Git, GitHub, SQLite, PostHog, PostgreSQL, Neovim, UNIX, Redis, REST, gRPC

Concepts: Low-Latency Systems, Concurrency, API Design, Distributed Systems

Education

Bachelor of Science (Honours) in Computer Science

• Expected: First, Penultimate year student

High School Diploma

Experience

• Drove a 97% reduction in KPI dashboard latency , directly improving data accessibility for business intelligence teams, by engineering a concurrent caching system in Python with per-query mutex locks
• Engineered a performant, company-wide KPI dashboard , consolidating data from 5 sources to provide BI insights
• Architected a highly scalable data platform , designing a caching system that enables easy integration of new data sources
• Analysed KPI trends using statistical methods, including lines of best fit and correlation analysis, to translate raw data into 20+ strategic business forecasts

Projects

• Engineered a custom Verlet integration physics engine , utilising JIT compilation to simulate 2,000+ agents at 60 FPS
• Implemented Spatial Partitioning via Cell Lists to optimise collision detection from O(N²) to O(N) ensuring performance
• Developed a Monte Carlo Search algorithm to optimise barrier placement, achieving a ~60% reduction in crowd pressure
• Integrated waveform analysis using Librosa to dynamically characterise agent behaviour via audio spectral data

• Designed a hybrid C++/Python engine, integrating a low-latency backend with an accessible interface for users to inject Numba-compiled machine code
• Iteratively enhanced engine strength by evolving the search algorithm from Minimax to Negamax with Alpha-Beta pruning , and implementing Quiescence Search to mitigate the horizon effect
• Developed a custom move generation engine and GUI from the ground up using C++ and SFML , utilising bitboard representations to optimise validation speed and minimise memory usage to be within 150 MB

• Architected a high-throughput, low-latency mock trading API in C++, engineered to handle 100+ concurrent users through asynchronous I/O and database connection pooling
• Engineered a low-latency data pipeline using gRPC and TimescaleDB, enabling millisecond-level queries on years of second-granular market data

• Built an interpreter for the A=B (a to b) esoteric language using Haskell, featuring a custom parser
• Achieved correct interpretation of 100% of official A=B examples, demonstrating deep understanding of idiomatic Haskell

Awards

• Acquired 9th out of ~200 teams (top 4.5% ) in Europe's largest and most competitive algorithmic trading hackathon (25% admittance rate)
• Forecasted financial market data with statistical and data analysis

• Awarded to 8% of entrants

Extracurricular

• Co-led the organisation of the largest Warwick Hackathon by liaising with sponsors and raising £10,000 in 2025
• Collaborated with 10+ societies and 7 companies across the Tech/Finance industries to deliver large-scale joint events
• Led the organisation of a large Chess Engine Competition , securing a sponsor from Optiver for the event and building the high-performance framework for the 40 competitors