High-Performance AI Computing Lab

We're Hiring!

The HAC Lab. at Kongju National University is actively looking for motivated undergraduate/M.S./Ph.D students.

Please contact me via cyoh@kongju.ac.kr with your CV if you are interested in joining our lab. (As of Jan. 2025)

Research Interest

The research topics that our lab is currently focusing on are mainly ...

• Parallel & Distributed Systems
  Modern computer systems feature heterogeneous massively parallel computing devices (e.g., GPU), which contribute to their exceptional computational capabilities. In addition, it is essential to break down AI tasks into smaller subtasks that can be processed concurrently on multiple machines for this large AI era. HAC Lab bridges the gap between needs of programmers and complexity of such modern computing devices in AI computing.
  Skills: CUDA, OpenCL, Hadoop, Spark, Pthreads, OpenMP, MPI, etc.

• On-Device AI
  One of the most important themes in AI is to deploy AI services on edge devices such as smartphone, tablet, wearable, and so forth. However, on-device AI also comes with challenges. Limited computing resources on devices may require model optimization and compression to fit within memory and power budgets, while meeting latency constraints. HAC Lab aims to achieve efficient-yet-accurate AI processing on edge devices.
  Skills: Android, TensorRT, TensorFlow Lite, Heuristic (e.g., Genetic Algorithm), etc.

• Cloud Computing
  High-performance AI computing involves the design and implementation of AI systems that can seamlessly expand their resources to handle growing workloads so as to ensure its efficiency. HAC Lab endeavors to deliver high-performance and efficient AI services by dynamically scheduling extensive user demands.
  Skills: Triton Inference Server, Kubernetes, Docker, Load Balancing, etc.