Who We Are Founded in 2019, Rock Bund Capital is a proprietary trading firm deeply committed to shaping the future of the cryptocurrency industry. We have an average daily trading volume reaching $1 billion and peak
具身智能大模型训练系统开发与优化实习生 上海 实习 职位描述 将会参与如下四个典型训练系统优化方向工作(包括但不限于下面四个方向)参与方向一:支撑大规模预训练/微调的高效稳定运行关键任务:1. 参与千卡级别分布式训练集群的框架优化,确保训练任务在大规模集群上的稳定性(任务失败率 xxx%)和可恢复性(断点续训时间 xxx分钟)2. 优化训练任务的吞吐量(Throughput),相比基线提升至少 20%3. 参与至少一种并行策略(数据并行、模型并行、流水线并行、MoE并行)的框架级实现或深度优化方向二:降低大模型训练显存占用,支持更大规模模型关键任务:1. 集成或优化至少一种显存节省技术(ZeRO-1/2/3、重计算(Activation Checkpointing)、混合精度训练),使相同GPU下的可训练参数量提升 30% 以上2. 参与FlashAttention、Flash-FFN等高性能算子在分布式训练框架中的集成与适配3. 验证并对比不同并行策略组合(如 FSDP + 张量并行)的显存效率与计算效率方向三:优化跨节点通信效率,降低通信开销关键任务:1. 使用 NCCL 或 CANN ACL profiling工具分析通信瓶颈(AllReduce、AllGather等),并提出至少 2 项有效优化方案2. 参与实现通信与计算的重叠(overlap)优化,使通信隐藏比例提升至 50% 以上3. 探索并验证低比特通信(如 FP8 梯度通信)在训练中的可行性与效果方向四:完善训练框架的可用性与可观测性关键任务:1. 开发或优化训练监控 Dashboard,覆盖多类关键指标(GPU利用率、内存占用、通信耗时、吞吐量、损失曲线)2. 参与实现训练任务的自动容错与恢复机制,支持节点级/进程级故障自动重调度3.
At Roche you can show up as yourself, embraced for the unique qualities you bring. Our culture encourages personal expression, open dialogue, and genuine connections, where you are valued, accepted and respected for who you are,
About Flexport: At Flexport, we believe global trade can move the human race forward. That’s why it’s our mission to make global commerce so easy there will be more of it. We’re shaping the future of
SummaryThe people here at Apple don’t just create products — they create the kind of wonder that’s revolutionized entire industries. It’s the diversity of those people and their ideas that inspires the innovation that runs through
We are Covestro. We are curious. We are courageous. We are colorful. We refine chemical material solutions with game-changing products. Let us empower you to push boundaries. Join us and our 18.000 colleagues now and together
GCIO - Asia & Middle East Technology Our GCIO organisation plays a critical role for the bank. This team partners with the businesses to build the platforms, systems, and products that our customers use every day.
Work ScheduleStandard (Mon-Fri) Environmental ConditionsSome degree of PPE (Personal Protective Equipment) required (safety glasses, gowning, gloves, lab coat, ear plugs etc.) Job Description When you are part of the team at Thermo Fisher Scientific, you’ll do
Understand, accept and comply with the related HSE responsibility commitment as part of daily work. MAJOR RESPONSIBILITIES Supervision • Maintains staff by recruiting, selecting, orienting, and training employees; maintaining a safe, secure, and legal work
Were seeking a creative and experienced Technical Program Manager to drive NVIDIAs Robotics initiatives in China. Youll own program execution across research, engineering, and product teams to ship robotics solutions at scale. This role is based
We are now looking for Chip Design Engineer for the Switch Silicon group. As a Chip Design Engineer at NVIDIAs Networking business unit, youll join a group of passionate engineers to design and implement the next
We are now looking for Chip Design Engineer for the Switch Silicon group. As a Chip Design Engineer at NVIDIAs Networking business unit, youll join a group of passionate engineers to design and implement the next
NVIDIA is seeking Software Performance Architects to optimize GPU kernel performance for state-of-the-art data-center platforms. We build automated, data-driven workflows to detect, explain, and prevent performance regressions across key deep learning workloads, partnering closely with kernel
As a Senior GPU & AI Infra Expert focusing on Cloud Service Providers (CSPs) in China, you will be a core technical pillar in NVIDIA’s CSP SA team, responsible for driving GPU/AI Infra technical strategy, system-level
Today, NVIDIA is tapping into the unlimited potential of AI to define the next era of computing. An era in which our GPU acts as the brains of computers, robots, and self-driving cars that can understand
NVIDIA is developing AI and Accelerated Computing solutions for the Automotive industry. We are looking for a Senior MLOps Manager to lead data operations, ground truth (GT) production and delivery. In this role, you will work
SummaryWe are looking for a passionate and experienced Software Engineer to help build next- generation of Applied Machine Learning Platform at Apple China. Applied Machine Learning Platform team provides backend services and infrastructure for various Machine
SummaryWe are looking for a highly motivated Software Engineer with 8–10 years of experience to join the ACM Platform and AML Services team. This role will design and build distributed backend services that power Apple’s fraud
SummaryAt Apple, new ideas have a way of becoming extraordinary products, services, and customer experiences very quickly. We believe that the diversity of our people and their ideas encourages the innovation that runs through everything we
Company Description WD is building the infrastructure behind the AI-driven data economy. As AI scales, so does data. Every interaction, every model, every system generates data that must be stored, managed, and made accessible over time.