具身大模型训练框架工程师/专家
Tasks
- Build large scale distributed training system
- Build large scale reinforcement learning post training system
- Build unified training framework for embodied intelligence
- Design checkpoint and fault recovery
- Develop high performance inference and automated evaluation system
- Enable topology aware scheduling and elastic scheduling
- Implement hybrid parallelism and optimization techniques
- Manage model versions and evaluation results
- Optimize cluster infrastructure for data IO and GPU compute
- Perform profiling and system performance optimization
Perks/Benefits
- N/A
Skills/Tech-stack
CUDA | Checkpointing | Collective communication | Computer Vision | Data parallelism | DeepSpeed | Distributed Training | Expert parallelism | FSDP | Fault Tolerance | Inference | Megatron-LM | Model Evaluation | Model Parallelism | Pipeline parallelism | Profiling | PyTorch | Python | Reinforcement Learning | Sequence parallelism | Tensor Parallelism
Education
Bachelor of Engineering | Bachelor of Science | Master of Science | PhD
Related jobs
-
高级数据工程师(座舱Ai) CNY 144K-240KApache Airflow | Apache Flink | Apache Hive | Apache Kafka | Apache SparkSenior-level Full Time北京6h ago
-
Senior-level Full Time武汉6h ago
-
Ai算法工程师 CNY 180K-360KComputer Vision | Continuous Delivery | Continuous integration | Data Analysis | Deep learningMid-level Full Time北京6h ago
-
车机 Ai 算法工程师(大模型 / 多模态 / 端侧部署) CNY 180K-360KC++ | Deep learning | INT4 Quantization | Inference engine | Int8 QuantizationMid-level Full Time北京6h ago
-
Mid-level Full Time北京7h ago
-
2027届秋招-Ai产品经理 CNY 144K-240KAgent Framework | Data Analysis | Language Models | Large Language Models | Model EvaluationEntry-level Full Time上海7h ago
-
AI Operations Builder CNY 144K-240KA/B | A/B Testing | API Integration | Agent workflow | AutomationSenior-level Full TimeHuangpu, Shanghai, China; Haidian, Beijing, China7h ago
-
None Full Time上海、成都、杭州8h ago
-
Principal Software Engineer(M365 Storage Fabric team) CNY 360K-510KAI-assisted engineering | AI-assisted engineering tools | Admission control | Automation | Autonomous Coding AgentsSenior-level Full TimeChina, Beijing, Beijing; China, Jiangsu, Suzhou; …18h ago
-
AI-assisted engineering | APIs | AWS | Algorithms | AutomationMid-level Full TimeChina, Beijing, Beijing; China, Jiangsu, Suzhou; …18h ago
-
Mid-level Full Time东莞2d ago
-
Mid-level Full Time深圳3d ago
-
具身数据策略与质控工程师 CNY 25K-37KBenchmarking | Computer Vision | Data Quality | Data Quality Management | Data cleaningEntry-level Full Time北京3d ago
-
Mid-level Full Time北京3d ago
-
数据Infra工程师 (流式) CNY 180K-300KApache Spark | Automated Data Labeling | Data labeling | Data pipeline | Deep learningMid-level Full Time北京3d ago
-
数据Infra工程师 (具身) CNY 180K-300KApache Spark | Deep learning | Distributed Systems | Error Handling | FastAPIMid-level Full Time北京3d ago
-
Senior-level Full Time北京3d ago
-
Mid-level Full Time北京3d ago
-
Mid-level Full Time上海3d ago
-
Avp, Mgoc Tno Analytics, Gra CNY 300K-420KApplication development | Dash | Data Analysis | Data Visualization | DocumentationContinuous professional development | Flexible working | Inclusive work environmentExecutive-level Full TimeGuangzhou, Guangdong, China R3d ago
-
Application Engineer - Associate CNY 25K-37KAgent workflows | Application development | Autogen | Git | LLM Application DevelopmentEntry-level Part TimeShanghai, China3d ago
-
Lead Data Scientist CNY 435K-510KAI Agents | AWS | Azure | CI/CD | Cloud ComputingHealthcare | Paid volunteering days | Retirement planning | Wellbeing initiativesSenior-level Full TimeCHN-Beijing-Zhong Guan Cun No.1, China4d ago
-
C# | C++ | CUDA | Computer Vision | GazeboCareer development | Hands-on experienceEntry-level Full TimeChina, Shanghai4d ago
-
Junior AI Full-Stack Engineer CNY 96K-144KAPI Development | Agent workflows | Audit Logging | Authorization | CachingEntry-level Full TimeSJ China Shanghai4d ago
-
CPU architecture | Compiler programming | Computer Architecture | Deep learning | GPU ArchitectureEntry-level Full TimeChina, Shanghai4d ago