Samuel Aboderin

Low-Latency C++ Engineer

MLH Fellow at G-Research, optimizing and rebuilding core components on Apache Arrow using C++, Cython, and Python. Computer Engineering student at the University of Lagos. I care about memory, cache behaviour, columnar data, and writing code the hardware respects.

Technical Skills

Languages
C++, Python, C, Cython, SQL
Systems
Memory layout & cache performance, lock-free data structures, SIMD, multithreading, low-latency design, columnar in-memory formats
Data & ML
Apache Arrow, PyArrow, NumPy, pandas, PyTorch, TDD
Infrastructure
PostgreSQL, Kafka, Airflow, Kubernetes, Docker, GCP
Networking
TCP/IP, UDP, socket programming, Wireshark, latency profiling
Tools
Linux, GDB, Valgrind, perf, CMake, Git, Bash

Education

University of Lagos, Nigeria
2022 – 2027
BSc. Computer Engineering
Relevant Coursework: Embedded Systems, Microprocessor Systems, Operating Systems, Computer Architecture, Digital Logic Design, Data Structures & Algorithms

Experience

G-Research · Software Engineering Fellow 2026 · Current
MLH Fellowship
  • Contributing to PyArrow, the Python bindings to Apache Arrow's C++ columnar in-memory data layer; changes reviewed and merged upstream into the Apache Arrow project
  • Work spans the full C++ → Cython → Python stack: implementing compute kernels in C++, authoring Cython bindings, and writing Python-level API and tests
  • Extended string compute functions to support large_utf8 via type-agnostic dispatch; improved null handling consistency across affected kernels
  • Mentored directly by the engineering team at G-Research, one of the leading quantitative research firms globally
↗ apache/arrow ↗ GitHub profile
Coach John Leadership & Community Engagement Initiative Jun – Aug 2025 · Remote
Software Engineer Intern · Canada
  • Engineered Python backend services using Django REST Framework, volunteer onboarding flows, donation tracking, and PostgreSQL schema design
  • Implemented data pipelines for synchronizing donor and volunteer data across third-party APIs, improving data consistency and operational efficiency

Projects

apache/arrow C++ · Cython · Python · Open Source
  • Active upstream contributor to Apache Arrow — the industry-standard columnar memory format used across production data systems and financial infrastructure
  • Contributions via the MLH Fellowship under G-Research: compute kernel extensions, Cython binding generation, null handling improvements, and test coverage
hft-simulator C++
  • Price-time priority limit order book with O(1) order insertion and cancellation
  • Matching engine simulating bid/ask spread dynamics and trade execution
  • Memory pool design for cache-efficient order storage
quant-cpp-playground C++
  • Low-latency data structures: ring buffers, lock-free queues, memory pools
  • Cache-aware memory layout patterns and SIMD-optimized numerical routines
  • Benchmarks comparing naive vs. optimized implementations across workloads
nanopore-basecalling-benchmark Python · PyTorch
  • Benchmarked LSTM and TCN architectures for DNA sequence basecalling using CTC decoding
  • Evaluated sequence accuracy and computational throughput across model configurations
  • Modular training and evaluation pipeline for basecalling models
pod5-accelerator Python
  • High-throughput pipeline for processing Oxford Nanopore POD5 genomic data files
  • 40%+ throughput improvement via parallel I/O and vectorized batch processing
Telos Python
  • Gradient-based steering for Markovian decision processes
  • Policy gradient methods with variance reduction for agent behaviour control
Multi-Agent RL Python
  • Mean-field approximation to tractably scale multi-agent RL to large N-agent systems
  • Reduces interaction complexity from O(N²) to O(N) via mean-field distribution

Certifications

Contact

GitHub LinkedIn Codeforces