Skip to content
Search jobs inside your AI agent. Claude, ChatGPT, Gemini, Cursor and Grok
Connect

Tech jobs

Job listings

18.7K jobs
Alibaba Group
3 views 2 days ago

阿里云智能-服务器软硬件系统架构与解决方案专家-杭州/北京

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Architects Alibaba Cloud's GPU/AI ASIC server systems: plans accelerator component introduction, validates performance, builds stability/monitoring/diagnostic tooling, customizes drivers, firmware and SDKs, and guides chip and rack-scale system design for AI training and inference. Core tech: GPU/AI ASICs, CUDA/ROCm, kernel drivers, server architecture.
Alibaba Group
3 views 2 days ago

阿里云智能-异构计算软硬件结合开发专家-杭州/北京

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud (Beijing/Hangzhou) seeks a heterogeneous computing expert to build software-hardware features for rack-scale AI supernode servers: performance profiling, container images, fault diagnostics, and next-gen AI infra planning on GPU/AI ASIC hardware. Core stack: CUDA, C++/Python on Linux, GPU architectures, PyTorch, Megatron/DeepSpeed, vLLM/SGLang.

AI DeepSpeed GPU TensorFlow vLLM
Sign in to see your match
Alibaba Group
3 views 2 days ago

阿里云智能-异构计算软硬件结合开发专家-北京/杭州

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Senior engineer on Alibaba Cloud Intelligence's heterogeneous computing team building hardware-software integration for rack-scale supernode GPU/AI ASIC servers — compute, interconnect, monitoring, fault-diagnosis and performance-profiling capabilities, plus next-gen AI infra planning. Core stack: CUDA, C++/Python, Linux, PyTorch, Megatron, vLLM/SGLang.

AI DeepSpeed GPU TensorFlow vLLM
Sign in to see your match
Alibaba Group
7 views 2 days ago

阿里云智能-AI Infra异构软硬件结合研发专家--上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Expert-level AI infrastructure role at Alibaba Cloud in Shanghai: develop and tune low-level GPU/AI-accelerator operators, AI compilers, Mega Kernels and quantization for large-model training and inference, plus design distributed multi-node architectures and explore frontier tech (MoE, automated kernel generation). Core stack: C++, Python, CUDA, PyTorch.

AI C++ CUDA GPU LLM +6 skills
 Sign in to see your match
Alibaba Group
6 views 2 days ago

阿里云智能-AI Infra异构软硬件结合研发专家-上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud's AI Infra team in Shanghai seeks an expert to develop and optimize GPU/AI-accelerator operators, AI compiler and Mega Kernel fusion, quantization, and distributed large-model training/inference architectures, working closely with chip and framework teams. Core stack: C++, Python, PyTorch, CUDA/CUTLASS/Triton, tensor and pipeline parallelism.

AI C++ CUDA GPU LLM +6 skills
 Sign in to see your match
Alibaba Group
7 views 2 days ago

阿里云智能-AI Infra视频媒体软硬件结合优化-深圳or上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Inference and post-training acceleration engineer on Alibaba Cloud Intelligence's AI Infra team, optimizing visual generation and multimodal (text/image/video) models across GPU/ASIC/FPGA hardware — quantization, attention, memory, compiler and multi-GPU parallel inference — using PyTorch, C++, Python and Rust. Based in Shenzhen or Shanghai.

AI CUDA Flux FPGA GPU +7 skills
 Sign in to see your match
Alibaba Group
4 views 2 days ago

阿里云智能-服务器超节点ScaleUp软硬件结合专家-北京/上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud expert role building software for Scale-Up switches and AI supernode cluster management: switch chip adaptation, telemetry, high availability, online upgrades, and automated benchmarking, plus tuning AI training/inference communication (collective comms, KV Cache) and next-gen supernode architecture. Core stack: C/C++, Python/Go, Linux, networking, distributed systems.

C++ InfiniBand RDMA
Sign in to see your match
Alibaba Group
3 views 2 days ago

阿里云智能-服务器超节点内存池化软硬件结合研发专家-上海/北京

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Expert systems R&D role at Alibaba Cloud Intelligence building unified pooled/tiered memory (HBM, CXL, host memory) hardware-software for AI supernode servers. Work spans Linux kernel memory management, high-speed interface drivers, memory pooling software, and performance tuning of AI (KV Cache) and datacenter workloads in C/C++ and Python.

C++
Alibaba Group
4 views 2 days ago

阿里云智能-服务器超节点ScaleUp软硬件结合专家-上海/北京

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud is hiring an expert to build Scale-Up switch software and supernode cluster-management platforms for AI infrastructure: switch chip adaptation, telemetry, HA and online upgrades, plus LLM training/inference communication tuning and next-gen supernode architecture. Stack: C/C++, Linux, networking, distributed systems.

C++ InfiniBand RDMA
Sign in to see your match
Alibaba Group
3 views 2 days ago

阿里云智能-服务器超节点内存池化软硬件结合研发专家-北京/深圳/上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud expert role designing the memory system for AI super nodes: building a unified tiered memory pool across GPU/NPU HBM, host memory, and CXL expansion memory. Day-to-day spans memory-pooling management software, kernel drivers/system software (hotplug, DMA, Linux memory management), and performance tuning for LLM training/inference using C/C++, Python, and Linux kernel internals.

C++ PCIe
Sign in to see your match
Alibaba Group
2 views 2 days ago

阿里云智能-服务器超节点内存池化软硬件结合研发专家-上海/北京/深圳

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

R&D expert for Alibaba Cloud's AI super-node memory systems: design unified tiered memory across GPU/NPU HBM, host and expanded memory, build memory-pooling management software and CXL/high-speed interface drivers, work on Linux kernel memory management, and tune performance for LLM training/inference and datacenter workloads. Requires 5+ years systems software experience with C/C++ and deep Linux

C++ PCIe
Sign in to see your match
Alibaba Group
3 views 2 days ago

阿里云智能-异构计算软硬件结合开发高级工程师-杭州/深圳

NewBe an early applicant APACSenior H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud senior engineer for AI hardware simulation and software-hardware codesign: builds parameterized models of chip compute, KV-cache memory hierarchy, interconnects (HBM, NoC, PCIe, CXL, NVLink/RDMA) and cluster topology to converge next-gen AI server and chip specs from model workloads and SLAs. Core stack: C/C++, Python, SystemC/ESL, performance modeling.

GPU PCIe
Sign in to see your match
Alibaba Group
4 views 2 days ago

阿里云智能-异构计算软硬件结合开发专家-杭州

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud is hiring in Hangzhou for an expert who builds performance profiling and diagnostics tooling for heterogeneous GPU/AI infrastructure — analyzing vLLM/SGLang/TRT-LLM inference frameworks, troubleshooting GPU stability (ECC/XID/NCCL, OOM), and optimizing multi-node cluster utilization with Python/C++/CUDA.

AI C++ DeepSpeed GPU LLM +3 skills
 Sign in to see your match
Alibaba Group
2 views 2 days ago

阿里云智能-异构计算软硬件结合开发高级工程师-深圳/杭州

NewBe an early applicant APACSenior H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Senior engineer on Alibaba Cloud's heterogeneous computing team building full-chain simulation for large-model inference: modeling request scheduling, KV cache, parallelism, and communication to characterize AI server latency, throughput, and capacity, and pushing the tech into mainstream inference frameworks. Core stack: C++, Python, SGLang/vLLM.
Alibaba Group
3 views 2 days ago

阿里云智能-异构计算软硬件结合开发专家-杭州/深圳

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Alibaba Cloud is hiring a heterogeneous computing software/hardware expert to optimize AI chip operators for large-model inference. The role architects the ZhenDan operator library, develops fused/quantized GPU/NPU kernels, integrates them into SGLang/vLLM inference frameworks, and profiles end-to-end performance bottlenecks with tools like Nsight and NCU.
Alibaba Group
6 views 2 days ago

阿里云智能-服务器软硬件结合开发专家-深圳/上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Works on software-hardware co-optimization at Alibaba Cloud to accelerate visual and multimodal generative AI models (text, image, video generation) on GPU/ASIC/FPGA — covering quantization, attention, memory and compiler optimization, and multi-GPU parallel inference using C++/Python/Rust with PyTorch, TensorRT-LLM, and TVM.

AI CUDA Flux FPGA GPU +7 skills
 Sign in to see your match
Alibaba Group
6 views 2 days ago

阿里云智能-服务器软硬件结合研发专家-深圳/上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Expert R&D engineer at Alibaba Cloud Intelligence accelerating vision/multimodal generative models (text, image, video) on GPU/ASIC/FPGA hardware — model quantization, attention/memory/compiler optimization, and multi-GPU parallel inference, working in C++/Python/Rust on PyTorch alongside hardware and algorithm teams.

AI CUDA Flux FPGA GPU +7 skills
 Sign in to see your match
Alibaba Group
11 views 2 days ago

阿里云智能-表格存储技术专家-杭州/北京/上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Technical expert on Alibaba Cloud's Tablestore team, building the distributed multimodal data platform that powers Agent memory and knowledge storage — designing vector/scalar hybrid storage, high-concurrency vector indexing and retrieval, multi-tenant metering and isolation, and production stability at PB scale. Core work in C++/Java with deep integration of AI frameworks like LangChain/LlamaInde

AI Cassandra CLI C++ Elasticsearch +10 skills
 Sign in to see your match
Alibaba Group
4 views 2 days ago

阿里云智能-服务器软硬件系统架构与解决方案专家-杭州/北京/上海

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

Expert role at Alibaba Cloud building and stabilizing AI compute systems: selects and validates GPU/CPU/network/storage hardware, manages firmware/OS delivery (BIOS, BMC, NIC/GPU firmware), and keeps massive (~10k-GPU) training and inference clusters stable, troubleshooting across firmware, drivers, and runtime.

GPU InfiniBand PCIe
Sign in to see your match
Alibaba Group
3 views 2 days ago

阿里云智能-大模型训练系统研发工程师-PAI平台

NewBe an early applicant APAC H-1B sponsor history — This employer has had H-1B petitions approved by USCIS in the last five fiscal years, per the USCIS H-1B Employer Data Hub. The approval belongs to the employer and is not a commitment to sponsor this role.
China

R&D engineer on Alibaba Cloud's PAI team building the one-stop AI platform behind the Tongyi large models: core development of the deep-learning training engine for MoE, multimodal (VLM/VLA) and Agentic RL workloads, boosting training throughput and stability across heterogeneous accelerators, plus monitoring and expert support for enterprise customers. Stack: Python/C++, PyTorch, Megatron/DeepSpe

DeepSpeed JAX RLHF
Sign in to see your match