Yun Kai Peng is a software engineer at Bloomberg, where he has spent the past three years working on the firm’s electronic trading solutions. He works in Modern C++, carefully balancing performance, reliability, and the occasional existential question about undefined behaviour. His current interests include multi-threaded application design and concurrency optimization for scalable, high-performance systems. Yun Kai Peng earned a bachelor’s degree in mathematics and computer science from McGill University.
Modern C++ graph-based libraries such as Taskflow provide efficient parallel execution by scheduling tasks across a pool of worker threads. In practice, many systems execute latency-sensitive tasks that should be prioritized over background processing. This work presents a design and implementation of priority-aware scheduling for C++ Taskflow that preserves its lightweight task-graph model while improving responsiveness under contention.