内容来源：https://news.microsoft.com/source/asia/features/from-prompts-to-partnership-how-ltms-rajesh-kumar-collaborates-with-microsoft-365-copilot/

内容总结：

全球科技服务公司LTM首席信息官拉杰什·库马尔表示，使用微软365 Copilot人工智能助手两年来，其提示词正变得越来越长、越来越复杂。最初他仅输入单句指令，如今已发展为先在OneNote上撰写成段说明，再交由Copilot处理。

库马尔指出，Copilot深度集成于Teams、Outlook等日常办公工具，能自动理解上下文，帮助他从海量邮件和聊天信息中聚焦关键事务。随着基础效率提升成为常态，他的使用重点已从优化工作流转向在全公司培育AI应用文化。

如今，他常借助Copilot的“研究员”智能体处理战略议题，例如筛选行业峰会案例、评估供应商等。这类任务需提交详尽提示词以获取综合分析报告，这减少了他对工作繁忙同事的依赖，为工作提供了高效起点。

作为公司AI推广负责人，库马尔团队正开展部门定制化培训，并面向业务用户举办黑客松活动，引导员工通过低代码工具Copilot Studio自主创建AI智能体。目前已有智能体能根据企业资源规划系统和员工数字简历，为项目快速匹配人才。

库马尔在个人生活中也频繁使用Copilot。近期在欧洲旅行时，Copilot为其推荐的夜间游览路线获得了出色体验。

中文翻译：

拉杰什·库马尔使用微软365 Copilot人工智能助手已有两年，在此期间，他输入的提示词变得越来越长。

最初，他只会输入简单的单句指令，例如："这封邮件是关于什么的？"或"写一封礼貌的拒绝邀请函。"

如今，他会先在OneNote上起草整段指示，再交给Copilot处理。

"提示词正变得越来越复杂，"全球科技服务公司LTM（前身为LTIMindtree）的首席信息官库马尔表示，"人机协作已发展到全新层次。"

起初，他对这款AI助手能通过Teams和Outlook等微软365产品套件融入工作流程感到惊叹。他利用它处理"如海般的邮件和聊天信息"，从而专注于需要关注的重要事务。

"最让我震撼的是微软将Copilot嵌入我们日常核心工具的方式。无需复制粘贴……它就能理解上下文。这正是我最初开始使用它的原因，那份惊喜感至今犹存。"他提到的Work IQ正是Copilot背后的智能层。

然而，这些效率提升很快成为常态。库马尔开始进一步探索——从仅将AI用于工作流程，转向在全公司培育AI使用文化。

对他而言，这意味着借助Copilot智能体"研究员"来获取所谓"战略议题"的洞察。这些议题范围广泛，既包括评估LTM的哪些案例研究最能展现其在行业峰会特定受众面前的实力，也涵盖针对员工效率工具潜在供应商的深度财务核查与市场调研。

相应的提示词也变得详尽。例如在替换现有软件时，他可能会向Copilot发出如下指令：

"该平台实施成本高昂且操作复杂。请全面总结IT服务行业的替代方案，对比所有产品优劣，并分析潜在风险与隐患。最终生成详细报告。"

使用"研究员"功能也意味着库马尔不再需要频繁打扰本就忙于本职工作的同事。

"过去我常会找同事询问'分享你的观点、研究这个课题、帮我整理幻灯片'，"他说，"这些无疑都是他们不愿被打扰的额外负担。现在处理这类工作时，我已经有了绝佳的起点——直接使用研究员智能体。"

作为首席信息官，库马尔还负责引领LTM全面应用人工智能。他的团队开始针对各部门开展Copilot专题培训，并为职能用户（而非软件开发者）组织黑客松活动。

目的何在？旨在向用户全面展示平台功能，包括用于构建AI智能体的低代码工具Microsoft Copilot Studio。

"下一阶段的发展核心是引入智能体思维，"库马尔说，"用户能自主创造什么？"

这些培训的成果之一，是能够通过企业资源规划系统与员工数字简历，为项目匹配具备相应时间与技能的LTM员工的智能体。

"智能体正在帮助我们为合适项目快速匹配最佳人才，"他表示。

库马尔在个人生活中也使用Copilot，最近一次是在度假期间。他于晚上7点抵达某欧洲城市后，便让Copilot推荐接下来三小时的最佳游览地点。

"它只推荐了几个地点，结果那些正是当晚全城最精彩的去处，"他回忆道，"我们度过了一段美妙时光。"

题图说明：LTM首席信息官拉杰什·库马尔。照片由LTM提供，背景由微软Copilot生成。

林艾琳通过微软资讯平台报道人工智能领域动态，重点关注AI如何提升亚洲民众的生活品质。她曾任新加坡《海峡时报》副国际新闻编辑，目前仍不定期撰写周末专栏。欢迎通过LinkedIn与她联系。

英文来源：

Rajesh Kumar’s prompts on Microsoft 365 Copilot have grown longer since he began using the AI assistant two years ago.
Then, he was posing one-liners like: “What is this email about?” and “Write a polite note declining this invitation.”
Now he drafts paragraphs of instructions on OneNote first, before handing them over to Copilot to work on.
“The prompts are becoming complex,” said Kumar, chief information officer at global technology services company LTM, previously known as LTIMindtree. “The collaboration has evolved to a different level.”
In the beginning, he marveled at how the AI assistant was woven into workflows via the Microsoft 365 suite of products, like Teams and Outlook. And he used it to cut through “the sea of emails and chats” so that he could focus on the issues that needed his attention.
“What struck me big time is the way Microsoft has embedded Copilot within the core tools that we use day in and day out. And there’s no need to copy and paste … it gets the context. So that is where I got into using it first and that excitement still follows,” he said, referring to Work IQ, the intelligence layer behind Copilot.
Soon however, these productivity gains became the new normal. Kumar explored further – shifting from using AI just for workflows to growing a culture of AI usage across the company.
For him, this meant turning to Researcher, the Copilot agent, for insight on what he calls “strategic topics.” These can range from evaluating which of LTM’s case studies would best showcase its talents for the particular audience at an industry summit to in-depth financial checks and market research on potential vendors for an employee productivity tool.
Not surprisingly, the prompts are extensive. To replace existing software, for example, he might prompt Copilot as follows:
“The platform is very expensive and complicated to implement. Create an exhaustive summary of alternatives available in the IT services industry, a comparison of all of these products and then provide what could be the risks and pitfalls. Generate a detailed report.”
Using Researcher also means Kumar no longer needs to lean heavily on his colleagues, who already have their hands full with their own projects.
“I used to go and ask them, ‘Share your perspective, research this, compile slides for me,’” he said. “I’m pretty sure those are disruptions and interruptions they never wanted. Now for any such work, I already have a great starting point. I just use the Researcher agent.”
As CIO, Kumar is also responsible for leading AI adoption across LTM. His team began conducting department-specific Copilot sessions and organized hackathons for functional users instead of software developers.
The purpose? To introduce users to the whole capability of the platform, including Microsoft Copilot Studio, a low‑code tool for building AI‑powered agents.
“The next phase of evolution really is introducing agent thinking,” said Kumar. “What can they create on their own?”
One outcome of these sessions is agents that match the availability and skills of LTM’s workers – via the enterprise resource planning system and their digital resumes – to a project.
“We have agents helping us to find the right talent for the right project which can be staffed the quickest,” he said.
Kumar also uses Copilot in his personal life, most recently on holiday. He had arrived in a European city at 7 p.m. and asked Copilot to suggest the best places to visit for the next three hours.
“It recommended just a couple of spots, and they turned out to be the best spots in the entire city at that time of night,” he said. “We had a beautiful time.”
Top image: Rajesh Kumar, CIO, LTM, Photo; LTM. Image background generated with Microsoft Copilot.
Lim Ai Leen reports on AI for Microsoft Source, focusing on how it’s improving lives in Asia. Ai Leen was formerly associate foreign editor at The Straits Times in Singapore and still pens an occasional weekend column. Contact her on LinkedIn.