{"code":0,"data":{"job_post_detail":{"id":"7646709717774960938","title":"顶尖应届-端云协同调度工程师（Agent OS）-MiClaw","description":"1、架构设计：定义分层架构（能力层 / 编排层 / 运行时层 / 适配层 / 可观测层），输出参考实现与技术白皮书；\n2、Agent 编排：实现高效的 DAG / Graph 调度、工具调用、Multi-Agent 协同，支持成本与延迟感知的调度策略；\n3、推理 Cache 优化：对接 vLLM / SGLang 等推理后端，针对 Agent 工作负载设计 Prefix Cache 复用、批调度；\n4、长时序执行：构建跨会话、分层式 Memory，以及面向长链路的检查点、重放、追踪与可观测组件；\n5、Prompt 工程化：沉淀 Prompt 版本管理、自动评测、回归测试工具链，推动 Prompt 自我演进。","requirement":"1、计算机 / 人工智能 / 数学等相关专业博士；有 Agent / LLM 系统 / RAG系统 / AI搜索等方向研究经历者优先；\n2、精通Python / Java / C++ / Rust 至少一门开发语言；具备 AI Infra 或系统软件一线研发经验；\n3、熟悉 LLM 训练 / 推理 / Agent 框架生态，具备从架构、业务流程到代码实现都基于 AI 原生范式设计能力；\n4、在 Agent 编排、推理优化、提示词工程、多智能体协作中至少一个方向有深入研究或落地经验；\n5、有顶会论文（NeurIPS / ICML / ACL等）或主流开源项目（vLLM / SGLang / LangGraph / MCP 生态等）贡献者优先。","recruit_type":{"id":"201","name":"正式","en_name":"Regular","i18n_name":"正式","depth":2,"parent":{"id":"2","name":"校招","en_name":"Campus","i18n_name":"校招","depth":1,"children":[],"active_status":1,"selectability":1},"children":[],"active_status":1,"selectability":1},"publish_time":1780388404106,"channel_online_status":1,"code":"A230240","job_id":"7646708993311082815","city_list":[{"code":"CT_11","name":"北京","en_name":"Beijing","location_type":3,"i18n_name":"北京","py_name":"beijing","mdm_code":"MDCY00000001","node_status":1}],"job_post_info":{"recruitment_type":{"id":"201","name":"正式","en_name":"Regular","i18n_name":"正式","depth":2,"parent":{"id":"2","name":"校招","en_name":"Campus","i18n_name":"校招","depth":1,"children":[],"active_status":1,"selectability":1},"children":[],"active_status":1,"selectability":1},"HighlightList":[],"JobChannelPublishList":[],"job_post_object_value_map":{},"address_list":[],"city_list":[],"correlation_job_list":[],"tag_list":[],"storefront_list":[],"target_major_list":[],"job_post_process_time_list":[],"job_level_id_list":[]},"city_info_list_for_delivery":[{"code":"CT_11","name":"北京","en_name":"Beijing","location_type":3,"i18n_name":"北京","py_name":"beijing","mdm_code":"MDCY00000001","node_status":1}],"tag_list":[],"storefront_mode":1,"storefront_list":[],"process_type":2},"recommend_job_post_List":[]},"message":"ok","error":null}