职位&公司对比
职位详情
- 北京
- 不限
- 本科
- CNN
- 编译器
- LLVM
- GCC
- AI
- 算法设计
- TensorFlow
Job Summary: We are seeking a highly skilled Senior Member of Technical Staff (SMTS) Compiler Software Engineer to join our team. In this role, you will be responsible for designing, developing, and optimizing compiler solutions that drive the performance and efficiency of our cutting-edge hardware platforms. You will collaborate with a dynamic team of engineers to push the boundaries of compiler technoloBOSS直聘gy, enabling innovative solutions for complex software challenges. Key Responsibilities: Design and implement advanced compiler optimizations and code generation techniques to maximize performance for AMD NPUs. Develop and maintain core components of the compiler, including front-end, midkanzhundle-end, and back-end. Analyze and optimize code performance, ensuring alignment with target platform requirements. Collaborate closely with hardware and software teams to define and implement features that enhance overall system performance. Conduct performance analysis, debug issues, and provide solutions for compiler-related problems. Stay up-to-date with the latestkanzhun advancements in compiler technology, programming languages, and hardware trends. Provide technical guidance and mentorship to junior engineers. Qualifications: Education: Master’s, or Ph.D. in Computer Science, Computer Engineering, or a related field. Experience: Minimum of 5+ ykanzhunears of experience in compiler development or a related field. Technical Skills: Proficiency in C/C++ and familiarity with modern programming practices. Deep understanding of compiler architecture, optimization techniques, and code generation. Experience with LLVM, TVM, MLIR, or other compiler toolchains. Strong background in performance tuning and debugging. Familiarity with hardware architectures such as GPUs, CPUs, or custom accelerators Soft Skills: Excellent problem-solving skills, strong analytical thinking, and effective communication skills. Preferred Qualifications: Experience with domain-specific languages (DSLs) or designing custom compiler extensions. Familiarity with machine learning frameworks and their optimization for hardware acceleration Knowledge of scripting languages such as Python for automation tasks. Prior contributions to open-source compiler projects.
职位详情
- 北京
- 3-5年
- 硕士
- 大模型推理
- 大模型优化
岗位职责 1. 配合模型应用和模型训练同事,上线大模型推理服务,推动大模型相关算法的落地; 2. 持续跟进前沿技术,优化大模型推理性能,提升线上模型服务吞吐量,降低模型推理延迟,控制推理机器成本 3. 开发和优化大模型推kanzhun理系统,提升模型推理系统的性能,稳定性,可观测性,易用性等 4. 使用GPU编程和系统优化手段实现高性能的推理优化算法,并将其落地于大模型推理系统中 岗位要求 1. 高性能计算,AI编译器,计算机科学,数值分析优化等相关教育背景,硕士及以上学历; 2. 熟练掌握Python,有工程开发经验,熟练掌握boss另一门系统级编程语言(如C++/Rust)者优先; 3. 熟悉常见的LLM推理加速框架和方法原理,如onnx/TensorRT/TensorRT-LLM/vllm/sglang,能够根据业务需求定制优化者优先; 4. 了解计算机体系结构,熟悉CPU/GPU下性能优化基本原理和方式方法,有服务端GPU程序加速经验者优先; 5. 具备CPU/GPU异构编程能力,包括但不限于CUDA, Triton,Candle等;熟悉至少kanzhun一种GPU加速库,如cublas,cutlass等; 6. kanzhun熟悉分布式推理常用加速方法,有分布式系统开发经验或大模型分布式部署经验者优先; 7. 具备严谨细致的工作态度,较强的团队协作能力
技能解析
- C/C++
数据来自CSL职业科学研究室
技能解析
- 开发和优化
- 优化算法
- 编程语言
- 熟悉分布式
- 相关算法
- 编程能力
- 系统开发
- CUDA
- 基本原理
- 工程开发
- 协作能力
- 系统优化
- 团队协作
- 开发经验
- 性能优化
- 团队协作能力
- 模型训练
- 分布式系统
数据来自CSL职业科学研究室
工作时间
公司福利
- 五险一金
- 补充医疗保险
- 定期体检
- 年终奖
- 股票期权
- 带薪年假
- 餐补
- 通讯补贴
- 交通补助
- 节日福利
- 零食下午茶
备注
职位发布者未明确表明公司信息,具体可咨询职位发布人进行确认。