大语言模型后训练/Agentic算法工程师
Tasks
- Build agent training and interaction environments
- Build large language model agent
- Conduct training and evaluation
- Create task orchestration and automated evaluation
- Design reward functions and harnesses
- Develop agentic reinforcement learning
- Develop tool systems and user simulators
- Drive LLM post training
- Implement multi tool calling
- Optimize multi turn reasoning and decision making
- Perform error analysis and scheme optimization
- Reproduce frontier post training methods
- Research and improve vehicle dialogue systems
- Set up data production and feedback loops
Perks/Benefits
- N/A
Skills/Tech-stack
Distributed Training | Function Calling | Java | Language Models | Language Processing | Large Language Models | Machine Learning | Memory | Model Evaluation | Natural Language | Natural Language Processing | On Policy | On policy Distillation | OpenRLHF | PPO | Planning | Policy Distillation | Preference Learning | PyTorch | Python | RLHF | RLVR | React | Reflection | Reinforcement Learning | Reinforcement Learning Framework | Reward Modeling | TRL | Tool-Calling | TypeScript | VeRL
Education
Bachelor of Engineering | Bachelor of Science | Master of Science | PhD
Regions
Countries
States
Related jobs
-
Entry-level Full Time北京 R3h ago
-
Senior-level Full Time北京3h ago
-
Nlp / Llm 应用工程师 CNY 180K-360KAgent | Anthropic API | Automated testing | Chroma | Dialogue ManagementEntry-level Full Time北京 R3h ago
-
Ai数据产品经理 CNY 240K-420KAgent Orchestration | BI | Data Governance | Data Warehouse | Dimensional ModelingMid-level Full Time北京3h ago
-
机器人VLA算法研究员 - XiaomiRobotics CNY 500K-500KAction Generation | Deep learning | Diffusion Models | Machine Learning | Mixture of ExpertsEntry-level Full Time北京3h ago
-
VLA训练infra算法工程师 - XiaomiRobotics CNY 240K-480KBF16 | C++ | CPU Memory Networking Fundamentals | CUDA | DeepSpeedMid-level Full Time北京 R3h ago
-
具身世界模型训练INFRA工程师 - XiaomiRobotics CNY 180K-360KAPIs | Deep learning | Distributed Training | Fault Tolerance | LinuxMid-level Full Time北京3h ago
-
具身世界模型数据工程师 - XiaomiRobotics CNY 96K-144KComputer Vision | Data Engineering | Machine Learning | Visual encodingMid-level Full Time北京3h ago
-
具身世界模型推理INFRA工程师 - XiaomiRobotics CNY 240K-480KCFG | Computer Vision | Diffusion Models | Expert parallelism | FP8Senior-level Full Time北京3h ago
-
具身智能算法工程师-模型 CNY 500K-500KActor-critic | Computer Vision | Deep learning | Distributed Training | Implicit Q LearningEntry-level Full Time北京 R3h ago
-
Mid-level Full Time北京3h ago
-
Agent 全栈研发工程师（前/后端）-MiMo CNY 180K-300KAPI Design | Agents SDK | Authentication | Authorization | AutogenMid-level Full Time北京3h ago
-
Mid-level Full Time北京 R3h ago
-
Entry-level Full Time深圳3h ago
-
Entry-level Full Time深圳、杭州3h ago
-
Mid-level Full Time上海5h ago
-
Senior-level Full Time上海、武汉、北京5h ago
-
算法工程师-大模型数据方向 CNY 240K-360KApache Spark | Clustering | Data Annotation | Data Deduplication | Data GovernanceSenior-level Full Time上海5h ago
-
数据开发工程师（Ai知识方向） CNY 180K-300KData Governance | Data Modeling | ETL | Elasticsearch | Information ArchitectureMid-level Full Time上海5h ago
-
Mid-level Full Time上海5h ago
-
Senior-level Full Time上海5h ago
-
Ai 应用研发工程师（上海） CNY 240K-480KAgent workflow | Docker | Embeddings | Evaluation and optimization | Function CallingSenior-level Full Time上海5h ago
-
大模型算法工程师（开放域对话） CNY 180K-300KData Deduplication | Data cleaning | DeepSpeed | Dialogue State Tracking | Distributed TrainingMid-level Internship上海、北京5h ago
-
Mid-level Full Time上海6h ago
-
数据产品经理（数据/Ai向） CNY 192K-300KClickHouse | Data Governance | Data Modeling | Data Quality | Data WarehouseSenior-level Full Time上海6h ago