AI Infrastructure Software Engineer — CosmosLab
Tasks
- Build and improve RL system data flow
- Build and improve distributed training backend
- Create and implement training infrastructure for AI foundation models
- Define reliability and efficiency metrics
- Develop and improve inference and evaluation pipelines
- Develop and improve pre training and SFT pipelines
- Improve training and RL workload efficiency scalability and resiliency
- Integrate simulation and robotics environments as RL environments
- Root cause triage and resolve system failures
Perks/Benefits
- N/A
Skills/Tech-stack
Activation checkpointing | C++ | CI/CD | CUDA | Checkpointing | Continuous batching | DTensor | Distributed Systems | Distributed Training | FSDP | Fault Tolerance | Fault Tolerant Restart | Fault-tolerant | Fine Tuning | Inference Serving | KV cache | Megatron | Mixed Precision | Parallelism | Preemption | Pretraining | PyTorch | Python | Reinforcement Learning | Reinforcement Learning Rollout | Sharding | Supervised Fine Tuning
Education
Related jobs
-
Mid-level Full Time北京18h ago
-
Mid-level Full Time北京18h ago
-
多模态数据算法工程师 CNY 216K-360KComputer Vision | Data Mining | Data Observability | Data cleaning | Deep learningEntry-level Full Time上海18h ago
-
具身智能数据算法工程师 CNY 144K-240K3D Gaussian Splatting | 3D Vision | Amazon S3 | Anomalous Data Detection | Apache ArrowMid-level Full Time北京18h ago
-
Mid-level Full Time深圳19h ago
-
Entry-level Full Time北京19h ago
-
多模态大模型算法工程师 CNY 240K-480KAblation Study | Agile | Computer Vision | Data Annotation | Data DeduplicationSenior-level Full Time北京19h ago
-
多模态大模型算法工程师 CNY 240K-480KComputer Vision | Data Augmentation | Data Deduplication | Data Mining | Data PreprocessingSenior-level Full Time深圳19h ago
-
多模态大模型算法工程师 CNY 240K-480KAgile | Computer Vision | Data Annotation | Data Deduplication | Data FilteringSenior-level Full Time上海19h ago
-
多模态大模型算法工程师 CNY 240K-480KAgile | Attribution Analysis | Automatic Labeling | Computer Vision | Data AblationSenior-level Full Time成都19h ago
-
Senior-level Full Time北京19h ago
-
Data Platform Engineering-Senior Principal, VP CNY 360K-600KAI Agent | AI Agent Development | AWS | Agent Development | Agent systemsEmployee networks | Flexible work/life support | Inclusive development opportunities | Paid volunteer daysSenior-level Full TimeHangzhou, China20h ago
-
Data Engineering - AI Platform Engineer CNY 300K-380KAWS | Access Control | Authentication | Azure | CI/CDEmployee networks | Flexible work/life support | Inclusive development opportunities | Paid volunteer daysMid-level Full TimeHangzhou, China20h ago
-
Entry-level Internship Part TimeWuxi, Jiangsu, China20h ago
-
Data Engineering - Agentic Engineer CNY 360K-540KAWS | Access Control | Apache Iceberg | Apache Spark | CI/CDEmployee networks | Flexible work/life support | Inclusive development opportunities | Paid volunteer days | Vibrant employee networksSenior-level Full TimeHangzhou, China20h ago
-
Mid-level Full Time深圳23h ago
-
Mid-level Full Time北京1d ago
-
Mid-level Full Time北京 R1d ago
-
大模型算法研究员-MiMo CNY 500K-500KActive Learning | C++ | Curriculum learning | Data Preprocessing | Deep learningEntry-level Full Time北京1d ago
-
Miclaw-端云协同调度专家 (Hybrid AI Architect) CNY 240K-360KCloud Computing | Consistency protocols | Distributed Systems | Edge Computing | Language ModelsHybrid workSenior-level Full Time北京 R1d ago
-
Senior-level Full TimeChangchun, CN1d ago
-
Embedded Software Engineer CNY 180K-300KBSP Debugging | Bootloader development | C# | C++ | CANAutonomy | Flexible work schedule | Healthy work-life balance | Mentoring | Paid vacationMid-level Full TimeShanghai, CN, 2012061d ago
-
Entry-level InternshipWuxi, CN1d ago
-
Big Data Engineer - Intern CNY 28K-50KAgile | Amazon EMR | Amazon S3 | Amazon Web Services | Apache AirflowEntry-level InternshipWuxi, CN1d ago
-
Big Data Engineer - Intern CNY 28K-50KAWS EMR | AWS S3 | Agile | Airflow | Apache HadoopFlexible clubs and activities | Mentorship opportunities | On-site meals | Team collaborationEntry-level InternshipWuxi, CN1d ago