포지션 상세
페블스퀘어는 PIM아키텍처를 기반으로 AI 반도체 설계와 AI 솔루션을 개발하는 팹리스 기업입니다. 폰 노이만(Von Neumann)구조의 한계를 극복하기 위해 고성능/초저전력의 PIM기반 AI 반도체를 성공적으로 양산하고, 멀티 코어 AI 반도체를 개발하여 국내외에서 실증 중이며 AI 반도체 상용화와 활용 촉진을 위해 다양한 AI 솔루션을 개발 중입니다.
This role focuses on designing and optimizing models for Analog Processing-In-Memory (PIM) architectures, covering the full stack from model design to hardware-aware optimization.
Key Responsibilities
• Design and develop AI/ML models optimized for on-device edge environments
• Research and implement model architectures tailored for Analog PIM systems
• Perform model optimization and compression (e.g., quantization, pruning)
• Optimize models under hardware constraints such as latency, power, and memory
• Collaborate with hardware teams on algorithm–hardware co-design
• Validate and evaluate models on real devices and simulators
• Explore and apply state-of-the-art techniques in edge AI and PIM technologies
• Hands-on experience with deep learning frameworks (e.g., PyTorch, TensorFlow)
• Strong understanding of model optimization techniques (quantization, pruning, distillation, etc.)
• Experience deploying models on edge devices (mobile, embedded, or IoT)
• Experience or solid understanding of hardware-aware AI design
주요업무
We are looking for a highly motivated engineer to develop AI models optimized for next-generation edge devices.This role focuses on designing and optimizing models for Analog Processing-In-Memory (PIM) architectures, covering the full stack from model design to hardware-aware optimization.
Key Responsibilities
• Design and develop AI/ML models optimized for on-device edge environments
• Research and implement model architectures tailored for Analog PIM systems
• Perform model optimization and compression (e.g., quantization, pruning)
• Optimize models under hardware constraints such as latency, power, and memory
• Collaborate with hardware teams on algorithm–hardware co-design
• Validate and evaluate models on real devices and simulators
• Explore and apply state-of-the-art techniques in edge AI and PIM technologies
자격요건
• MS degree or higher in Computer Science, Electrical Engineering, or a related field (or equivalent experience)• Hands-on experience with deep learning frameworks (e.g., PyTorch, TensorFlow)
• Strong understanding of model optimization techniques (quantization, pruning, distillation, etc.)
• Experience deploying models on edge devices (mobile, embedded, or IoT)
• Experience or solid understanding of hardware-aware AI design
