Description
As part of NVIDIA's Analytics and Data Intelligence (ADI) group, this team develops libcudf,the open-source CUDA C++ library that accelerates database and DataFrame operations.
The team is looking for an outstanding engineer and/or scientist to apply their parallel programming skills to accelerate open-source software libraries for GPU-based data processing.
In this position, you will drive speed-of-light performance in structured data processing, spanning hardware from single workstations to multi-node GPU supercomputers.
Responsibilities:
- Own development for UcxExchange in Velox (30%)
- GPU-to-GPU communication in Presto
- Drive the feature roadmap, improve performance, and ensure correctness
- Optimize multi-node performance for analytical queries with Presto GPU (30%)
- Targeting both on-prem clusters and public cloud servers
- Lead projects of interest in Presto, Velox, cuDF (40%)
- Such as operator completeness, GPU memory oversubscription, efficient CPU fallback
- Plus your ideas
Requirements:
- 8+ years of experience in Computer Science or Software Engineering
- MS degree or PhD in computer science, engineering, or a related field, or equivalent experience
- Strong Modern C++ programming skills
- You care deeply about robust, readable, high-performance code
Nice to Have:
- Expertise in high-performance communication protocols (e.g., UCX, NCCL) and distributed algorithms for query engines
- Familiarity with RAPIDS cuDF
- Experience in distributed workflow development and debugging
- Passion for open-source software development and publishing your work in technical blogs and conferences