박광현 연구실

High-performance, AI-centric data platforms that unify diverse data and enable intelligent data infrastructure at scale. Research includes vector databases, GPU-accelerated systems, cloud-native architectures, and data lakehouse systems.

System infrastructures optimized for AI workloads across diverse environments, bridging algorithm design and real-world deployment. Includes scalable training infrastructure, inference systems, and resource management.

Leveraging ML to build self-optimizing systems with learned optimization, predictive scheduling, and adaptive management. Includes ML-based query optimization and AutoML for system configuration.

Domain-focused, data-centric AI applications leveraging large language models and advanced AI techniques. Includes Korean language models, medical AI, and social science applications.

New paradigms for large-scale AI models with focus on accuracy, efficiency, and scalable deployment. Includes LLM optimization, diffusion models, and multimodal foundation models.

Co-optimizing algorithms, software, and emerging hardware for data- and AI-intensive workloads. Includes GPU, TPU, NPU, DPU, and CXL integration.

Exploring quantum computing to unlock new capabilities in data management and machine learning. Includes hybrid quantum-classical systems and quantum algorithms.