Name: Google 6小時提示工程課程，10分鐘讓你搞懂！ (Google's 6 Hour Prompt Engineering Course in 10 Minutes)
Uploaded: 2026-01-22T23:51:55.000Z
Duration: 10 min 55 s

程度副詞

I just finished Google's six hour prompt engineering course and it is hands down the best AI training I have ever taken.

我剛上完 Google 的六小時 prompt engineering 課程，這絕對是我上過最棒的 AI 訓練。

So I'm saving you the time and I've packed every essential tactic Google revealed into this guide so you can master the entire system in under 12 minutes.

所以我幫你省下時間，把 Google 公開的所有關鍵技巧都濃縮在這份指南裡，讓你能在 12 分鐘內搞懂整個系統。

情態副詞

Of course, I can't fit every lesson from this nine hour course into one video, so if you want the complete training and an official certificate from Google you can actually put on your resume or LinkedIn.

當然，我沒辦法把這九小時的課程全部塞進一支影片，如果你想要完整的訓練和 Google 的官方證書，讓你真的能放在履歷或 LinkedIn 上。

I'll leave a link to Google's course in the description.

我會在說明區放上 Google 課程的連結。

It starts with understanding how the model actually thinks.

首先要了解模型實際是如何思考的。

So Google structures their entire course around five core task context references.

所以 Google 把他們的整個課程架構在五個核心任務的上下文參考上。

Evaluate and iterate the foundation of everything is task.

評估與迭代，一切的基礎就是任務。

This is simply what you want the AI to do.

Not the general topic, but the exact output you need.

不是籠統的主題，而是你需要的確切輸出。

A bad task is help me with email, while a good task is reformatting the sentence to write an email to my gym staff about a schedule change.

爛的任務是「幫我處理一下 Email」，而好的任務是「把這句話改寫成一封 Email，寄給健身房的員工，告知他們行程有變動」。

But here's where Google takes it further.

You can add two elements to make the task even stronger.

你可以加入兩個元素，讓任務更強大。

When you tell the AI to act as an expert, you aren't just playing pretend.

當你告訴 AI 扮演專家時，你不是在假裝。

You are priming the model to access a specific set of vocabulary and logic.

你是在引導模型存取特定的詞彙和邏輯。

adding the instruction like act as a physical therapist ensures you get safety tips and anatomical focus rather than just a list of exercises.

加上「扮演物理治療師」這樣的訓示，能確保你得到安全提示和解剖學重點，而不只是一份運動清單。

It changes the entire vibe of the answer.

Second is format, and this is your biggest time saver.

第二個是 format，這能幫你省下最多時間。

If you don't define the format, the AI defaults to a generic wall of text.

如果你沒有定義格式，AI 會預設成一堆文字。

By asking for a bulleted list, a markdown table, or even a JSON snippet.

透過要求條列式清單、Markdown 表格，甚至是 JSON 片段。

You force the AI to organize its thinking.

You stop getting raw information that you have to fix later and start getting usable deliverables that are ready to.

你就不會得到需要之後再修改的原始資訊，而是能直接使用的成果。

These two additions turn a basic task into a specific structured result.

這兩項補充能將基本任務變成具體的結構化結果。

Once the task is clear, you need to layer in context.

一旦任務清楚了，你就需要加入內容。

Context is the background data that steers the model.

The more information you provide, the less the AI has to guess.

你提供的資訊越多，AI 需要猜測的就越少。

If you just type write landing page copy for my website, you get generic text that could apply to anything.

如果你只打「寫我的網站的登陸頁文案」，你會得到適用於任何內容的通用文字。

But look at what happens when you inject context.

但看看加入內容後會發生什麼。

I'm building a project management tool for freelance designers.

我正在為自由接案的設計師打造一個專案管理工具。

The Target users are 25 to 40 and they are frustrated with tools like Asana being too complex.

目標用戶是 25 到 40 歲，他們對 Asana 這類太複雜的工具感到沮喪。

My product focuses on visual timelines and client portals.

我的產品著重於視覺時間軸和客戶入口網站。

Now the AI knows exactly who you are talking to and the output becomes targeted instead of generic.

現在 AI 就知道你在跟誰說話，輸出的內容也會變得更有針對性，而不是通用的。

Google's course emphasizes this repeatedly.

They assume the AI will figure it out, but it won't.

他們以為 AI 會自己搞定，但其實不會。

Now, to really lock in the quality, you add references.

現在，要讓品質真正到位，你就要加入參考。

References are examples that show the AI what you're aiming for.

參考就是給 AI 看的範例，讓它知道你的目標是什麼。

Sometimes words aren't enough to capture a specific vibe or structure.

有時候光用文字沒辦法傳達出特定的感覺或結構。

Let's say you are writing a product description and you need it to match your specific brand voice.

假設你正在寫產品說明，而且你希望它符合你特定的品牌風格。

Paste in three of your best descriptions and tell the model.

把你最好的三個說明貼進去，然後告訴模型。

Write a new description using the same style as these examples.

寫一個新的說明，使用跟這些範例一樣的風格。

Or, if you are creating social media content, feed it your top performing posts and tell the AI to analyze why they worked.

不然，如果你正在製作社群媒體內容，就餵給它你表現最好的貼文，然後告訴 AI 去分析為什麼它們會成功。

Then have IT generate new posts that follow that exact pattern.

接著讓它產生新的貼文，而且要完全模仿那個模式。

References turn vague instructions into concrete targets.

參考範例能把模糊的訓示變成具體的目標。

It stops the model from guessing your style and forces it to match what you already know.

這樣模型就不會亂猜你的風格，而是會強迫它去符合你已經知道的東西。

Once you get a response, you move to evaluate.

當你收到回應後，接著就要進行評估。

This means checking if the output actually hits the mark.

這代表要檢查輸出的內容是不是真的打到靶心。

It sounds basic, but this is where most people fail.

聽起來很基本，但這就是大多數人失敗的地方。

They skim the text, settle for good enough, and move on.

他們只是快速掃描文字，覺得差不多就好，然後就繼續往下。

You need to actively verify that the output matches the task, hits the right tone, and

你需要主動確認輸出的內容有沒有符合任務、語氣對不對，

你提問、檢查、調整，然後再提問。

Google provides four specific ways to fix a broken prompt.

Google 提供了四種具體的方法來修正有問題的 prompt。

The first method is to simply revisit the framework.

第一種方法是直接重新檢視框架。

Go back to the start and check if you missed the context or forgot to assign a Persona.

回到起點，檢查你是不是漏掉了情境，或是忘了指定 Persona。

Sometimes the fix is just filling in the gaps you missed the first time.

有時候，修正的方法只是補上你第一次漏掉的資訊。

If that doesn't work, try the second method.

如果這樣還是不行，就試試第二種方法。

Breaking into simpler sentences the AI processes information just like a human does.

把資訊拆解成簡單的句子，AI 處理資訊就像人一樣。

If you dump a massive paragraph of instructions, it gets overwhelmed.

如果你丟一大段指令給它，它會不知所措。

Don't write a run on sentence about a Q1 strategy targeting Gen Z with budgets and KPIs all in one breath.

不要寫一個又臭又長的句子，把 Q1 策略、目標客群 Gen Z、預算和 KPI 全部一口氣講完。

Break it down and write something like Create a strategy for Q1 target Gen Z, include a budget, add KPIs.

把它拆開來，寫成「為 Q1 制定一個目標是 Gen Z 的策略，包含預算，再加上 KPI」。

Writing the same information but with clearer structure results in better output.

用更清晰的結構寫同樣的資訊，輸出的結果會更好。

The third tactic is to use analogous tasks.

第三個技巧是使用類似的任務。

If the direct approach fails, try a different angle.

如果直接的方法失敗了，就試試不同的角度。

If write a business proposal is giving you dry, boring results, switch the frame.

如果寫商業提案的結果很乾、很無聊，就換個框架。

Ask it to write a persuasive argument for a partnership.

請它寫一個有說服力的合作夥伴關係論點。

You are changing the mental model the AI uses, which often gives a much better result.

你正在改變 AI 使用的心智模型，這通常會帶來更好的結果。

If you ask for video ideas and get generic results, clamp down on the requirements.

如果你要求影片點子卻得到很籠統的結果，那就收緊要求。

所以告訴它必須在 90 秒內完成。

Now the AI has to work within a box, which makes the ideas specific instead of broad.

現在 AI 必須在框架內工作，這會讓點子變得具體而不是籠統。

Now beyond text, there is multimodal prompting.

現在除了文字，還有多模態 prompt。

This is where models like Gemini really separate themselves from the pack as besides just reading text, they can process images, audio and video natively.

這就是像 Gemini 這樣的模型真正脫穎而出的地方，因為除了讀取文字之外，它們還能原生處理圖像、音訊和影片。

Say you are redesigning a website and need feedback.

假設你正在重新設計一個網站，需要一些回饋意見。

Instead of wasting time describing the layout in words, just upload a screenshot.

與其花時間用文字描述版面配置，不如直接上傳一張螢幕截圖。

Then command it to analyze this homepage design.

然後指令它分析這個首頁設計。

Identify three specific areas where user attention might drop off and suggest improvements.

找出三個使用者注意力可能下降的特定區域，並提出改進建議。

Or if you are a musician working on a track, upload the audio file and ask it to describe the mood of this piece.

或者，如果你是一位正在製作歌曲的音樂家，上傳音訊檔案，請它描述這首曲子的情緒。

Then suggest five alternative directions.

The framework still applies here you are just replacing vague text descriptions with high fidelity visual or audio references.

框架在這裡仍然適用，你只是用高傳真的視覺或音訊參考，取代模糊的文字描述。

But we need to address the elephant in the room.

但我們需要談談最明顯的問題。

字幕列表影片播放

Google 6小時提示工程課程，10分鐘讓你搞懂！ (Google's 6 Hour Prompt Engineering Course in 10 Minutes)

specific

entire

massive

phrase