如何善用AI工具提高工作效率

— 常用AI工具的演示与培训

AI时代的来临改变了职业人员的工作方式，如何适应新的行业趋势? 如何在五花八门的AI工具中选用最适合的工具来提高工作效率？怎样有效地收集信息、分析数据，执行任务？

通过亲自实践和比较，四位来自不同行业的职业人员根据不同的场景和需求，精心设计了这期AI工具培训班，带您从用 chatbots 提升到使用 agents 以及多个 agents 的协同工作。

在讲员演示之后，您会有机会动手练习，快速掌握几个常用的AI工具: Google AI (NotebookLM, Gemini), Claude (MCP, Skills), 和 OpenClaw。

注册费: $70/人 (3/7之前), $100/人 (从3/7起)。为保证质量，学习班人数将限制在 10-20人（10人以上开班，不超过20人）。报名费将用于讲员报酬和协会活动经费。

时间：3/14 周六,  3-6 PM 美西 ｜ 6-9 PM 美东 | 3/15 周日, 7-10 AM 中国

主持人：

戴兰青，在金融支付公司担任首席技术项目经理，曾在微软，百度和浪潮等全球领先的公司工作，主导多个跨职能项目，推动战略计划和创新解决方案的落地。我对人工智能的潜力充满热情，目前利用业余时间专注于学习人工智能及其在实际项目中的应用，希望通过学习与实践，不断探索如何将这一技术与业务需求相结合。

第一部分：AI for Research

敬华博士在结构生物、生物信息、数据管理等领域深耕20多年，近年来在业余时间持续关注人工智能在生物医药行业中的应用，在25年初成立了博雅 AI 俱乐部，26年初并入CAST-SD。做为一位科研人员，她将分享如果用 NotebookLM 来阅读文献、综合资料、做图做表，做视频内容总结和会议总结。

Thomas Li 博士毕业于柯朗数学科学研究所，获数学博士学位。其研究领域主要集中于随机控制与优化、机器学习、以及数值分析。本次讲座将概述一套严谨的框架，用于将 Google Gemini 整合至科学研究中。与会者将探讨如何通过与人工智能的迭代式交互，来综合分析学术文献、论证数学推导，并将理论模型转化为算法计算机程序。本次演讲包含一项综合演示，详细展示如何针对深度学习模型生成、调试及优化Python代码。最终，与会者将掌握一套系统的方法论，以实现从理论分析到计算实现的无缝衔接。

第二部分：AI for Data Analytics

Kelvin He 是资深 AI 与数据科学从业者，专注于构建机器学习框架与 AI 驱动的解决方案。他擅长端到端模型开发，能够将业务挑战转化为可落地的数据策略，并长期关注 AI 基准评测、优化技术与大模型（LLM）应用，帮助企业做出更有把握的技术决策。

在本次分享中，Kelvin 将现场演示 Claude 如何从“对话助手”升级为真正能落地的生产力工具，重点聚焦 MCP（Model Context Protocol）与 Skills。通过真实案例，你将看到 MCP 如何让 Claude 连接数据与工具，实现可追溯、可引用来源的分析与问答；以及 Skills 如何把团队最佳实践固化为可复用流程，用于自动生成报告/简报、标准化洞察输出与高一致性结果交付。整场演示将把抽象指令转化为具体行动与可交付成果，显著节省日常知识工作与数据分析的时间成本。

第三部分：AI for Automation

曹国梁是北卡三角区人工智能协会 (RTAI) 的会长和联合创始人，该协会是一家致力于推广和促进人工智能创新的非营利组织。凭借其计算机科学和软件开发方面的专业知识，他带领 RTAI 成功举办了两届大型会议，并通过二十多场聚会和技术讲座，建立了一个蓬勃发展的社区，这些活动和讲座吸引着众多热情参与者。

曹国梁将演示 OpenClaw（原名 Clawdbot）——近几个月来涌现的最引人注目的人工智能产品之一。与典型的聊天机器人或传统代理不同，OpenClaw 是一款真正的人工智能助手，能够主动代表您执行任务。通过现场演示，您将看到它如何无缝集成到日常工作流程中，将抽象的指令转化为具体的操作，从而节省大量重复性工作时间。

Speaker Biographies and Program Information in English

Moderator: 

Lanqing Dai, I work at a FinTech company as a principal technical program manager, and I have previously worked at global leading companies such as Microsoft, Baidu, and Inspur, where I led multiple cross-functional projects and drove the implementation of strategic plans and innovative solutions. I am passionate about the potential of artificial intelligence and am currently using my spare time to focus on learning AI and its applications in real-world projects, hoping to continuously explore how to combine this technology with business needs through study and practice.

Part 1: AI for Research

Dr. Holly Jing has been working on structural biology, bioinformatics, and database management for over 20 years. In recent years, she has been paying close attention to the applications of artificial intelligence in the pharmaceutical industry in her spare time. She founded Boya AI Club in early 2025, which merged into CAST-SD in early 2026. From the perspective of a scientific researcher, she will demonstrate how to use NotebookLM+Gemini to read literature, synthesize information, create graphs and tables, and summarize video contents and conference proceedings.

Dr. Thomas Li accomplished his Ph.D. studies in mathematics at the Courant Institute. His primary research focuses on stochastic control and optimisation, alongside machine learning and numerical analysis.

This session outlines a rigorous framework for incorporating Google Gemini into scientific research. Participants will explore how to engage with AI iteratively to synthesise academic literature, substantiate mathematical derivations, and translate theoretical constructs into algorithms and computer programmes. The presentation features a comprehensive demonstration of generating, debugging, and optimising Python codes tailored for deep learning models. Ultimately, attendees will acquire a systematic methodology to transition seamlessly from theoretical analysis to computational implementation.

Part 2: AI for Data Analytics

Kelvin He is a seasoned AI and data science professional specializing in building machine learning frameworks and AI-driven solutions. He leads end-to-end model development by translating business challenges into data strategies, and he is passionate about AI benchmarks, optimization techniques, and real-world LLM applications that help organizations make confident technology decisions.

In this talk, Kelvin will demonstrate how Claude can move beyond “chatting” into measurable productivity—especially through Model Context Protocol (MCP) and Skills. Through live demos, you’ll see how MCP connects Claude to real tools and data sources, and how Skills turn best practices into repeatable workflows for analytics and knowledge work. The session will showcase practical scenarios such as grounded Q&A with internal context, automated reporting/brief creation, and consistent insight generation—transforming prompts into traceable outputs and concrete actions.

Part 3: AI for Automation

Guoliang Cao is the co-founder and president of Research Triangle AI Society (RTAI), a nonprofit fostering AI awareness and innovation. Drawing on his computer science and software development expertise, he’s led RTAI to host two major conferences, and build a thriving community through dozens of meetups and tech talks that consistently draw enthusiastic crowds.

Guoliang will demonstrate OpenClaw (formerly Clawdbot)—one of the most compelling AI products to emerge in recent months. Unlike typical chatbots or conventional agents, OpenClaw functions as a true AI assistant that actively executes tasks on your behalf. Through live examples, you’ll see how it seamlessly integrates into daily workflows, transforming abstract commands into concrete actions and saving hours of routine work.