研究科学家-基础设施系统实验室(北京/杭州)
Core
Research and develop Approximate Nearest Neighbor (ANN) search algorithms, focusing on filtering retrieval, hybrid retrieval, and disk-based indexing schemes to optimize scalability, latency, and memory usage.
Role type
Senior Research Scientist (Infrastructure Systems)
Builds
Production-grade vector search systems and retrieval infrastructure
Domain
Search & Retrieval, Machine Learning, Distributed Systems
Deliverable
production ML models
Required skills
ANN algorithms, Quantization techniques, Graph indexing, Partitioned indexing, System-level performance analysis, C++ or Python
Preferred skills
Vector database construction, FAISS/ScaNN/HNSWLib/DiskANN frameworks, Distributed system architecture, GPU-accelerated retrieval, Hybrid/Multi-modal/LLM retrieval
Technologies
C++, Python, FAISS, ScaNN, HNSWLib, DiskANN
Responsibilities
Research and develop ANN search algorithms; Optimize algorithms for scalability and latency; Collaborate with engineering teams for prototype and production integration; Publish academic papers or contribute to open-source projects; Track advancements in vector retrieval and RAG systems.