For professionals, effective time management can help us use our time more efficiently and help us achieve work-life balance. Now, we can harness the power of AI to help achieve this goal.
Preface
Hello, I'm Zhang Wenjing.
When it comes to time management, remember the last time it became a hot topic was because of the time management guru surnamed Luo.
In a blink of an eye, four years passed like this.
So, is time management just as important for us ordinary people?
The answer is self-evident.
As a working person, the pace of work and life is fast, and many times because of busy work and neglect family and life.
Even though each of us has the same 24 hours of time every day, by skillfully planning these hours, we can create infinite value in a limited amount of time.
Therefore, mastering time management skills can not only improve our time efficiency, but also help us achieve a harmonious coexistence between work and life, develop good habits, and ultimately achieve our life goals.
Many people may think that time management is very troublesome, and it may not be necessary to have no time and energy for themselves.
But have you ever thought that we can manage our time with the help of the power of AI?
AI, as the existence that knows humans best outside of humans, after a series of trainings, may even know you better than you do.
In the process of experiencing many AI tools and large models, I found that AI tools such as Agent are the best suited for this scenario. (For example, AlphaGo is the best agent in the Go circle)
So what is an agent?
Agent = LLM + Planning Ability + Memory + External Tools.
Relying on the underlying large model, plus short-term and long-term memory, and finally human-computer interaction through external tools that are convenient for three parties to use.
Agent looked at a lot of kinds, including KIMI, byte bean bag, Zhipu Qingyan and other agents, and finally found that the better landing was DingTalk.
Because DingTalk itself is office software, it naturally has richer scenes, and then through the combination of Agent + scenes, the experience I feel is the best.
Combining the intelligence of the agent with my time management method, the results are eye-catching;
It solves many of my time management challenges and always schedules time for me at critical moments.
Sometimes, I even feel like it knows me better than I know (ORZ... ）
Today, I will talk to you about how to manage time efficiently and how to use AI to create a dedicated time management assistant.
Time management is really time-consuming
If we do our own time management, it's really a lot of energy.
Not to mention whether it is necessary for everyone to manage time, it takes a long time just to do time planning and allocation, let alone implementation and review.
Why? Now let's talk about time management.
Time management, is it really just time management?
Actually, no, time management is just a means, time management is mainly: ability management, energy management, and finally time management.
At the heart of energy management is the block of time.
For example, split time into 30-minute, 1-hour, and 2-hour time blocks.
Divide the time of the day into a relatively fixed number of units for use.
2 hours of time blocks to do things that require long hours of concentration;
The 1-hour block requires a certain amount of time to focus, but it can be completed within 1 hour as much as possible;
30 minutes to do some fragmented work.
The time block can be learned by the AI assistant, and let him help us split it according to this logic.
What is the core of managing time. In fact, it is to focus on the present.
At all times, know what you should be doing at the moment.
I usually give myself a panel like this, break down the time blocks, and determine the to-do for each day.
This can be done by the AI assistant and let them generate it for us.
The core of capability management is to find the pattern.
Everyone's work is essentially followed by a pattern;
Through pattern recognition, finding his pattern becomes a standard, then solidified, and finally refined.
It's like a product manager who has just learned a model or method, and when he uses it for the first time, it is very slow and needs to be practiced before he can digest it.
When you use it for the second and third time, you are very proficient and gradually standardize, and the speed will be very fast.
This is because we have found patterns and our abilities have improved.
What should I do before it becomes a standard?
For those tasks that cannot be completed as standard, there should be enough buffer and buffer time.
That way, at least your schedule won't be disrupted by a task.
So, here's the problem.
How to find a pattern, how to allocate time, how to predict the time of each task is a difficult problem.
In addition, the time management process is very long, taking me as an example, I iterated on the time management SOP for two years as follows:
Even though I've been doing this for 2 years, there are still things in this process that take me a lot of time to sort out:
- According to the monthly goal, break it down to a weekly task to-do list (120 minutes per week)
- Sync your weekly task to-do to your calendar and determine the exact time for each task. (60 minutes per week)
- Record and summarize the actual completion of each week (30 minutes)
- Weekly review and adjustment of the scheduled time for future tasks (30 minutes)
In the past, I used to predict the time through my own feelings, and then plan and schedule by myself.
In terms of time, at the end of a month, it is 600 minutes (120 minutes for a monthly plan dismantling + 2x (60 + 30 + 30) minutes for a four-week plan and review).
Today I found out that it took me 10 hours to increase the entropy of time management in the previous month!!
Recently, after researching Agent, AI helped me predict, fast and accurate.
When I discovered that DingTalk AI Assistant can directly generate AI-generated content in DingTalk, I began to try to build a "private customized time management assistant" for myself with prompt words.
to generate a review of these "bells and whistles" of these "bells and whistles" in my diagram below.
After I started using DingTalk's AI assistant, I counted the time spent on my time management from about 600 minutes per month to less than 100 minutes now, which is at least 6 times more efficient, and the effect is much better than what I organized by myself.
Students who have never used AI may think:
It's hard for AI to feed knowledge, right?
Prompt words are hard to do, right?
Is it complicated to configure parameters?
Isn't the AI feedback good enough......
From my experience, the effect can be played and can fully meet the needs of my daily work and life.
And building a DingTalk AI assistant is not so complicated, from configuration, adjustment to final use, it takes 10 minutes before and after, and it can be done in 3 steps~
Quickly set up an AI assistant in 10 minutes, once and for all
Step 1: Work backwards from the actual pain points to the AI work goals
First of all, clarify the work goal of the AI assistant: to improve my management efficiency according to my time management pain points.
Time spent on key points:
1. According to the monthly goal, break it down to a weekly task to-do list (120 minutes per week)
Goal 1: Be able to automatically help me plan my weekly to-dos
2. Sync to the schedule and determine the specific time for each task. (60 minutes per week)
Objective 2: To be able to predict when I plan to complete each task
3. Record and summarize the actual completion of each week (30 minutes)
Goal 3: Be able to help me review and give me some objective suggestions.
4. Weekly review and estimate the planned time for future tasks (30 minutes)
Objective 4: Task estimation is not very accurate, and there is no external auxiliary optimization mechanism.
I even hope that the AI will tell me which tasks are too long to complete and need to be trained and learned
Therefore, the tasks and objectives of the AI assistant are:
- Assist me in completing the decomposition of monthly tasks and generate weekly task plans;
- Determine my weekly task schedule and sort out the text content that needs to be synchronized to the daily to-do.
- Based on historical data, predict when I plan to complete each task;
- Assist me in completing weekly and monthly reviews, and provide me with reasonable advice;
If you find this step troublesome, you can directly use the DingTalk AI assistant market to find the [Help you write the role setting] AI assistant, and ask him to help you complete or further optimize your prompt words until you are satisfied.
Finally, I sorted out the prompt words I use now:
Step 2: Configure the DingTalk AI Assistant
When configuring the DingTalk AI assistant, the most critical step is to carefully set the role prompt words.
In the first step, we have prepared these prompt words.
Now, all we have to do is copy and paste these prompts into the character settings.
At the same time, you can test the feedback results of the AI assistant in real time on the right side of the interface.
This instant feedback mechanism allows you to visually see if the information returned by the AI assistant meets your expectations.
Based on the AI assistant's feedback, adjust the prompt on the fly until it performs exactly what you want.
With such a personalized configuration, your AI assistant will be able to better understand your needs.
For example, I send my monthly plan to the AI assistant.
Let him generate it to me in the following format:
The AI assistant will automatically generate my weekly schedule for the next month.
What surprised me the most was that AI could also recognize the nature of tasks!!
Remind me that some events don't have to be allotted hours!
I never expected to be able to recognize it here!
And the AI assistant predicts the planned start time and end time of each task for me according to my requirements~
Some friends may be curious, is the AI assistant's time prediction accurate?
Accuracy depends a lot on how we shape and train it.
As long as you have the heart to make the AI accurate, it will become accurate.
We want a personal time management assistant, an AI assistant that can be customized, is your AI assistant.
How to customize it privately?
What is used is the agent's long-term memory, uploading knowledge files, and letting the large model learn.
There is also this function in DingTalk, when configuring the AI assistant, you can upload a knowledge file, and the AI will automatically learn the content of the file, so that he understands you better.
For example, I feed him the time management data I have been doing for 2 years and let him predict the timing of my future tasks based on it.
We humans sometimes can't remember the events of 2 years ago, but AI remembers them clearly!
Judging from the predicted results, I am still quite satisfied!
Step 3: Build your AI assistant matrix
Think about it, if you have one assistant to plan your schedule, do you need more assistants to manage your time?
The answer is yes. I added a dedicated review assistant to myself.
It is precisely because DingTalk AI assistant supports DingTalk knowledge base, online documents, and local documents, which are ways to memorize knowledge for a long time.
So I "fed" all my personal habits to this assistant, including my definition of each task, my personal understanding of certain tasks, the usual free time, and so on.
In addition to this, I also told it about the completion of historical tasks, and it will learn and analyze it, and then predict a more reasonable planning time in future mission planning.
Therefore, such a snap of the process data, feed it to the AI assistant, and you let him learn it.
There will be new data later, and I will continue to supplement him, so that he can become an evolving intelligent partner for me, helping me learn from past experiences and do a more personal review for me.
Finally, let's take a look at the effect of letting the review assistant specify the review time for the time period:
Tell me how to arrange my time reasonably when I set a certain task in the future:
A lot of times, what I didn't think about, AI took care of for me......
I also have a habit of syncing the week's to-do to my schedule every week.
Here I've added another schedule assistant to sync your to-dos, just tell him what you're going to do, and the week's schedule panel will be generated quickly.
I have to say that DingTalk is very capable of playing through his endogenous ecology and with AI.
summary
In the past two years, I have been constantly refining my time management methods, and although I have a set of SOPs, it takes at least 600 minutes to do this time management every month.
With the advent of AI, it takes less than 100 minutes to do some simple tuning and training in a month, which is at least 6 times more efficient.
The barrier to entry for time management has become lower.
Through the DingTalk AI assistant, I made a time management assistant, a personal review assistant, and a schedule assistant, and let them help me complete:
- undefined 月度计划拆解
- undefined 每周任务待办
- undefined 每周复盘
- undefined predicts the timing of the to-do schedule
With these AI assistants, you're no longer alone, but a team.
When I learned about agents, the first thing that came to my mind was the time management agent.
Each agent is built on a powerful underlying model that understands the environment and makes sound planning and forecasting.
These agents are also capable of possessing human-like long-term memory capabilities through knowledge bases.
This long-term memory, combined with the short-term memory of the agent itself, results in an efficient and friendly way of interacting.
The way people interact with information is changing little by little with the times.
Let's go back and look at the information connectors of the past dynasties, from the ancient matchmakers and recommenders to modern newspapers, publishing houses, and telephone bureaus, to the navigation websites and search website information flow in the early days of the Internet, and then to the current self-media ecology such as Douyin and Xiaohongshu.
Looking to the future, what we can see is that in the era of AI, intelligent agents may set off a wave of information revolution just like 25 years ago when the Internet arrived.
Finally, the barrier to entry for AI is already low, and even more so for tools like DingTalk AI Assistant.
If 25 years ago, Internet people needed a sense of network, then at this moment, what is more important is a sense of intelligence.
I hope that you can use agents, try agents, and make agents.
Agents are good aids when used well, just like these assistants of mine.
By contacting the agent, gradually adapt to the way the agent interacts to improve the prompt ability.
It's like people who used the Internet a lot back then thought more quickly about the combination of their own industry and the Internet.
If you don't want to be abandoned by the times, it is easier to think about the various ways to land the agent by using more agents now.
If AI is a form of magic, there will be two types of humans in the future: wizards and muggles.
Prompts are spells, and agents like DingTalk AI assistants are magic wands.
If you want to become a magician and start experiencing the agent agent, I believe you can think about more scenarios like me.
This article was posted by @张文靖 on Everyone is a Product Manager and is not allowed to be reproduced without permission.
Image from Unsplash, based on the CC0 license
The views in this article only represent the author's own, everyone is a product manager, and the platform only provides information storage space services.