9小时谷歌提示词课程精华
YouTube原视频是将Google学习时长约9小时的提示词课程内容,浓缩到了20分钟,然后这篇文章是20分钟浓缩视频的笔记。。
YouTube原视频是将Google学习时长约9小时的提示词课程内容,浓缩到了20分钟,然后这篇文章是20分钟浓缩视频的笔记。
可以先了解Google的提示词课程讲了什么,你是否适合学习,或者里面某一点是你想要了解的,然后再去看看系统的课程。
原视频链接:https://www.youtube.com/watch?v=p09yRj47kNM&t=735s
Google提示词课程:Google Prompting Essentials | Coursera
课程分为四个板块:
模块一:像专业人士一样开始编写提示(Start writing prompts like a pro)
这个板块主要是介绍写提示词的基础框架
核心框架:TCREI五步法
1.任务(Task)- 明确目标:清楚表达你希望AI执行的具体操作
基础版: "suggest the gift related to anime for my friend's birthday"
添加角色设定:"Act as an anime expert to suggest an anime gift for my friend's birthday."
指定输出格式:"organize that data into a table"
- 上下文(Context)- 提供背景:上下文越丰富,输出质量越高
"Act as an anime expert to suggest a get related to anime for my friend's bthday and she is turning 29 years old."
参考资料(References)- 给出范例:当难以用语言描述需求时,提供具体示例最有效
评估(Evaluation)- 检查结果:获得输出后,评估是否符合预期
迭代(Iteration)- 持续优化:提示工程是循环改进的过程,很少一次到位
博主的记忆技巧: "Tiny Crabs Riding Enormous Iguanas"
四种迭代优化方法
当基础框架解决了80%的问题,剩余20%可通过以下方法解决:
- 重新审视框架 (Revisit the framework)
提供更多参考资料、例子、背景信息
完善角色设定
- 拆分复杂提示词 (Separate into shorter sentences)
将长篇提示词分解为简短、清晰的句子,像与普通人交谈一样,避免信息过载。
- 采用故事化表达 (Try different phrasing or analogous task)
用生动的故事替代枯燥的描述,让AI更好理解需求。
对比示例:
❌ 正常版:写一个产品营销文案
✅ 故事版:描述这个产品如何改变一个忙碌上班族的生活
- 引入合理限制 (Introduce constraints)
通过约束条件缩小范围,获得更精准的结果。
示例: 制作公路旅行歌单 → 限制为"90年代摇滚,适合夜间驾驶,情绪偏忧郁"
博主的记忆技巧: "Ramen Saves Tragic Idiots"
多模态提示词
现代AI模型不仅限于文本交互,还支持多模态输入与输出,包括图片、音频、视频和代码。
尽管交互媒介不同,但五步提示设计框架依然适用,只是需要更清晰地指明输入/输出的类型和相关背景。
图片输入: 输入一张新设计的美甲系列图片,要求AI撰写一篇有趣、简短的社交媒体推广文案。
"write a social media post featuring this image and then attach your nail art collection as a reference the post should be fun short and focus on the fact that's a collection of new designs I'm selling"

音频输入: 输入一段音乐,要求AI根据其“氛围”和“感觉”来创作短篇故事的场景和细节。
混合输入: 输入品牌的标志和颜色,要求AI创建一个数字预告片来推广活动。
使用 AI 工具存在两个主要问题:
1.幻觉 (Hallucinations)
当生成式 AI 工具提供的输出不一致、不正确或甚至是胡言乱语
- 偏见 (Biases)
LLM在人类内容上进行训练,因此也纳入了人类的偏见,如性别和种族偏见。
解决方案: 课程建议采用“人工干预”(Human in the Loop)的方法,始终检查并核实AI工具提供的内容,确保最终产出的准确性也是用户自己的责任。
……
模块二:设计日常工作任务提示(Design proms for everyday work task)
这个板块有很多演示,可以看下原视频或课程
电子邮件撰写:一位健身房经理需要撰写一封电子邮件,通知员工新的排班表更改
“I'm a gym manager and we have a new gym schedule write an email informing our staff the new schedule highlight the fact that the MWF Monday Wednesday Friday Cardio Blast class changed from 7:00 a.m. to 6:00 a.m. make the email professional and friendly and short so that the reader can skim it quickly here's the new schedule and you can actually attach the link that contains the new schedule.”

这份提示词包含了各类详细信息,比如:要求 AI 扮演的角色(健身房经理)、具体任务(撰写关于新排班表的邮件)、突出信息点(周一/周三/周五的有氧运动课从早上 7:00 改到 6:00),以及期望的输出风格(专业、友好且简短,易于快速浏览)
避免模糊描述,使用具体的语气指导:
❌ "写得随意一些"
✅ "用友好易懂的语气,就像向好奇的朋友解释一样"
更多技巧: 上传你过往的邮件作为语气参考样本
以下是其他案例展示:


……
模块三:数据分析与PPT(Data analysis and presentations)
数据分析注意事项:谨慎对待输入到AI中的数据,不要违反隐私政策,不要将敏感数据输入大型语言模型。
课程提供了一个针对杂货连锁店数据集的示例:
"Attached is a Google sheet of store data how can I create a new column in sheets that calculates the average sales per customer for each store."

然后进一步的分析:
"give me insights into the relationship between daily customer count items available and sales based on the given data."

AI不仅能提供具体操作指导,还能发现数据中的隐藏趋势
以下是视频中展示的关于PPT的提示词示例


……
模块四:使用AI作为创意或专家伙伴(Use AI as a creative or expert partner)
介绍了四种高级提示技巧:
- 提示链(Prompt Chaining):引导生成工具完成一系列相互关联的提示,逐步增加复杂性
小说营销案例:
第一步:生成书籍摘要

第二步:创建宣传标语

第三步:制定推广计划

- 思维链提示(Chain of Thought):要求AI逐步解释其推理过程,类似数学老师要求解释作业步骤
在提示中添加"解释你的思考过程",帮助理解AI推理过程,可以帮助AI改进决策
- 思维树提示(Tree of Thought):像树一样,同时探索多个推理路径和分支。适用于抽象或复杂问题
"imagine three different designers are pitching their designs to me all designers will write down one step of their thinking then share it with the group then all experts will go on to the next step Etc if any expert realizes they're wrong at any point then they leave the question is generate an image that's visually energetic and features images of art supplies in computers show me three suggestions in very different styles from simple to detailed and complex"

- 元提示 (Meta Prompting):元提示指的是当你陷入困境,不知道该使用什么提示词时,可以使用 AI 来帮助你生成提示词
……
AI 代理 (AI Agents)
模块四的最后一部分是关于AI 代理
AI Agents就像专家一样,用于帮助完成任务和回答问题, 可以有编码Agent、营销Agent、高尔夫Agent,甚至是朋友Agent
- 模拟代理(Agent Sim):可以模拟各种场景,如面试或角色扮演
HR培训示例:
"act as a career development training Simulator,
your task is to help interns Master interview skills and conduct conversation with potential managers,
"you need to support the following types of conversations articulating strengths and skills communicating professionally and confidently And discussing future career development goals once an intern has picked a conversation topic provide details about the situation in the interviewer's role then act as the interviewer and allow the intern to participate as the employee make sure to guide a conversation in a way that will allow the intern to exercise their interview skills,
continue to roleplay until the intern replies with jazz hands after the intern gives the stop rule jazz hands provide them with key takeaways from the simulation and skills they can work on"

- 专家反馈代理(Agent X):就选择的任何主题提供反馈,类似个性化导师或顾问
跑车公司提案反馈示例:
"you're my potential client the VP of advertising at a world famous sports car company known for its Innovation performance and Engineering Excellence.
you're considering hiring a creative agency to develop a new campaign that will attract a younger generation of buyers you're in a meeting with me the design director of a creative agency that's pitching a new campaign for your company.
act as my potential client when I provide answers critique the answers if needed ask follow-up questions continue the conversation until I give the stop roll break then give me a summary of the whole conversation highlighting ways I can improve my pitch.
I've included the brief the car company provided me that has all the relevant information for this project use the information from this brief to inform your answers."

创建AI Agent的五步框架
角色设定:定义专业身份和专长领域
背景信息:尽可能多地提供关于场景和对话的背景信息和细节
交互规则:设置对话或互动方式规则
停止机制:一个停止短语来结束对话
反馈系统:确保对话结束后有总结和建议
