大模型训练调优工程师
Tasks
- Analyze training stability and convergence performance
- Build multimodal model training pipeline
- Collaborate with algorithm team on model and system bottleneck optimization
- Design training quality evaluation and anomaly monitoring
- Develop parameter efficient finetuning methods
- Implement mixed precision and performance optimization techniques
- Implement multimodal data loading and coordination
- Improve training system performance using compiler optimization
- Maintain pretraining finetuning evaluation deployment workflows
- Manage multi GPU resource scheduling and DDP acceleration
- Optimize training pipeline multi task scheduling
Perks/Benefits
- N/A
Skills/Tech-stack
Adapter | Anomaly Detection | CI/CD | DDP | Data loading | DeepSpeed | FSDP | LoRA | Mixed Precision | Model Training | Multi-GPU | PEFT | PyTorch | Training evaluation | Transformer
Education
Bachelor of Engineering | Bachelor of Science | Master of Science | PhD
Roles
AI | AI Engineer | Engineer | Learning Engineer | Machine Learning Engineer
Regions
Countries
States
Related jobs
-
Technical Lead, Machine Learning CNY 144K-240KBias Evaluation | Cost Optimization | DPO | Data Pipelines | Evaluation PipelinesSenior-level Full TimeChina R9d ago
-
Staff Machine Learning Engineer CNY 144K-240KBias Evaluation | Cost Optimization | DPO | Data Pipelines | DeploymentSenior-level Full TimeChina R9d ago
-
Senior Software Engineer (RAG Backend Developer) CNY 277K-370KA/B | A/B Testing | ABAC | Access Control | Audit LoggingSenior-level Full TimeGuangzhou, Guangdong, China R10d ago
-
Alerting | Benchmarking | Data Pipelines | Distributed Systems | GPU ComputingMid-level Full TimeChina R13d ago
-
Mid-level Full TimeChina R13d ago
-
Entry-level Full TimeChina R14d ago
-
Principal Machine Learning Engineer CNY 144K-240KApache Arrow | Apache Spark | DPO | Deep learning | DeepSpeedSenior-level Full TimeChina R14d ago
-
AWQ | AWS | Accelerate | Azure | BatchingSenior-level Full TimeGuangzhou, Guangdong, China R20d ago
-
Senior-level Full Time上海 R20d ago
-
Mid-level Full Time上海 R20d ago
-
Senior-level Full Time上海 R20d ago
-
Data Infra高级研发工程师(AI与具身智能数据平台方向) CNY 144K-240KAsynchronous processing | Automated testing | CI/CD | CRD | Cloud NativeSenior-level Full Time上海 R20d ago
-
Ego-Centric视觉感知算法工程师 CNY 180K-300K2D Segmentation | 3D Gaussian Splatting | 3D Hand Pose | 3D Reconstruction | 6D PoseMid-level Full Time上海、北京 R20d ago
-
数据算法开发工程师/专家 CNY 240K-480K2D Computer Vision | 3D Computer Vision | ASR | Ablation Study | Active LearningSenior-level Full Time上海 R20d ago
-
Mid-level Full TimeGuangzhou (DTC), China R21d ago
-
Entry-level Full Time上海 R26d ago
-
None Full Time深圳 R30d ago
-
Ai 院--多模态团队--多模态理解算法研究员-强化学习方向 CNY 240K-480KDPO | Data Processing | Data pipeline | DeepSpeed | Distributed TrainingSenior-level Full Time北京 R1mo ago
-
Entry-level Full Time北京 R1mo ago
-
Ego-Centric 视觉感知算法工程师- Genie业务部 CNY 144K-240K3D Gaussian Splatting | 3D Pose Estimation | 3D point cloud | 6D Pose Estimation | Bundle adjustmentMid-level Full Time上海 R1mo ago
-
优才-具身大模型训练框架工程师-觅蜂子公司 CNY 500K-500KC++ | CPU Optimization | CUDA | CUDA graph | Communication overlapMid-level Full Time上海 R1mo ago
-
Platform Operations Engineer CNY 306K-364KALCOA plus | API Management | Alerting | CI/CD | Canary ReleasesRemote workMid-level Full TimeShanghai, CN R1mo ago
-
API Design | AWS | Agent Loop | Agent Orchestration | Async workflowsSenior-level Full TimeShenzhen, Guangdong Province, China - Remote R1mo ago