尚旭春
Employers on the program
- 2026TokenSpeed 独立贡献者1 talk
Events1
Conferences1
Talks1
Years active2026
The year around now
Talks in the past 12 months and announced talks ahead, one row per conference, January to December. Hover a mark for the talk.
Appearances by year
- AICon 全球人工智能开发与应用大会 (AICon Global AI Development and Application Conference)
Talks
| Date | Event | Conference | Role | Title | Links |
|---|---|---|---|---|---|
| Aug 21, 2026 | AICon 全球人工智能开发与应用大会 | 深圳 | AICon 全球人工智能开发与应用大会 (AICon Global AI Development and Application Conference) | speaker | 面向 Agentic 负载的下一代 LLM 推理引擎设计实践 as TokenSpeed 独立贡献者 随着 AI Agent 从 Demo 逐步走向生产环境，LLM 推理负载正在发生根本性变化。相比传统聊天场景，Agentic 工作负载具有高并发、短请求、多轮工具调用、逐 Token 生成等特点，推理引擎的优化目标也从追求极限吞吐，转向在保证低延迟的同时实现更高的 GPU Token 输出效率。…Show less随着 AI Agent 从 Demo 逐步走向生产环境，LLM 推理负载正在发生根本性变化。相比传统聊天场景，Agentic 工作负载具有高并发、短请求、多轮工具调用、逐 Token 生成等特点，推理引擎的优化目标也从追求极限吞吐，转向在保证低延迟的同时实现更高的 GPU Token 输出效率。 与此同时，模型架构和硬件平台也在快速演进。以 DeepSeek、Kimi 等新一代模型为代表，MoE、MLA、异构并行逐渐成为主流；Blackwell 等新一代 GPU 又引入了 TMEM、tcgen05 等全新的计算能力。这意味着，仅依靠 Kernel 优化已… |
Frequent co-speakers
No one else spoke at two or more of the same events.
Speakers with the same name are merged into one entry, so this page may combine more than one person. Affiliations are shown as each conference listed them.