- 地点
- 科隆，德国
- 类型
- 全职
- 资历
- junior
职责
- 扩展和改进主机编译器和运行时，以便在新型和新兴硬件平台上高效执行编译后的模型
- 协调 CPU、GPU 和 NPU 的执行，释放异构系统的全部性能潜力
- 在 MLIR 编译器与基于 C 的运行时之间工作，塑造编译模型的表示和执行方式
- 设计并实现基于 MLIR 的下一代编译器的新组件
- 与不断壮大的开源社区合作，围绕 Roofline 构建生态系统
必备要求
- 具备 C 和 C++ 的知识和实践经验
- 对编译器概念有理论或实践接触
- 计算机科学/电气工程或相关领域的学位
- 扎实的软件工程原理、数据结构、设计模式和算法知识
- 渴望与不断壮大的开源贡献者社区合作
- 具备解决复杂问题的雄心和好奇心，责任感强，能主动承担任务并推动解决方案，具备协作精神
优先条件
- 具备使用 MLIR 或 LLVM 的经验
- 熟悉 Vulkan、CUDA、Metal 或 OpenCL 等框架
- 具备人工智能算法知识，有使用人工智能加速器或编译到相关硬件的经验
- 掌握数据运动网络、数据流编程模型和优化人工智能库的相关知识
- 了解 CPU、GPU 或 NPU/加速器的硬件架构
- 具备 Python 开发经验
Responsibilities
- Extend and improve our host compiler and runtime for efficient execution of compiled models on new and emerging hardware platforms
- Coordinate CPU, GPU and NPU execution to unlock the full performance potential of heterogeneous systems
- Work at the boundary between our MLIR compiler and C-based runtime, shaping how compiled models are represented and executed
- Design and implement new components of the next-generation MLIR-based compiler
- Collaborate with the growing open-source community to build an ecosystem around Roofline
Must Have
- Knowledge and hands-on experience with C and C++
- Theoretical or practical exposure to compiler concepts
- Degree in Computer Science/Electrical Engineering or a related field
- Solid knowledge of software engineering principles, data structures, design patterns and algorithms
- Eagerness to collaborate with a growing open-source contributor community
- Ambition and curiosity to solve complex problems, strong sense of responsibility, ability to take initiative and drive solutions, and a collaborative spirit
Preferred
- Experience with MLIR or LLVM
- Familiarity with frameworks such as Vulkan, CUDA, Metal or OpenCL
- Knowledge of AI algorithms and experience with AI accelerators or compiling to related hardware
- Knowledge of data movement networks, dataflow programming models and optimized AI libraries
- Understanding of CPU, GPU or NPU/accelerator hardware architectures
- Python development experience
公司简介
Roofline 正在构建一个软件开发工具包，用于在边缘端的高性能硬件上运行任何模型。在 RooflineAI，我们助力您实现梦寐以求的边缘人工智能产品。我们的使命很简单：将芯片的强大功能带给创新者，并让他们能够更轻松地在顶尖硬件上使用尖端人工智能模型。