Okay, look, I don’t know how to kick this one off. This could’ve been titled One Weird Trick The Secret Cabal Of Productive Engineers Doesn’t Want You To Know About. Could’ve been, because it’s true (kinda - I haven’t checked in with the cabal in a while), but it also is a lame title.
Hey there! Great article, it resonated with me on a deep level 😂
You're absolutely right, but I have to admit that I've never been good at planning, especially for a completely new project. Creating a plan for a "simple task" like the one you mentioned is relatively easy to do.
However, when it comes to building something more complex, like a digital menu app, it becomes a different story. There are numerous factors to consider: you have different user roles, which requires authentication, authorization, and so on. You'll need at least one backend and one frontend, if not more. The example you provided in the article quickly becomes just a small piece of a much larger puzzle, and that always intimidates me. I worry that maintaining the plan will consume too much of my time.
Personally, I'm a fan of tools and always on the lookout for the best ones to use. So when I start creating a plan for a project like this, I begin jotting down ideas. But in the end, I always find myself thinking, "This takes too much time. How amazing would it be if there were a tool, especially in this era of AI, that could handle planning for me?"
I should clarify that by planning, I don't mean simply creating a to-do list. That's relatively easy nowadays with chat GPT 😉.
By planning, I mean being able to answer a fundamental question: How long will it take to build this?
Believe me, getting an answer to this question is like finding the Holy Grail 😂
To obtain an answer, you need to know the number of people involved, their availability (are they also working on another project?), the order in which tasks should be done, and who should do them. Some tasks may be dependent on others, and so on. Planning involves numerous factors and requires a significant amount of time to maintain.
In this AI era, I believe such time-consuming tasks should no longer be a necessity.
Trust me, I've spent countless hours trying to find something useful, but my searches have never led me to anything meaningful or, at the very least, something that works the way I envision it.
Here's an approach I think could work:
Tasks should be represented as a nested list, where each task can have subtasks. Dependencies and priorities can be defined for tasks. Each task should also have an estimated effort associated with it (expressed as a range, e.g., 8-12 hours) and the required skill set (or even a list of skills). Additionally, individuals should be defined, with assigned skills and skill levels (ranging from 1 to 10 to represent junior to senior), along with their available working hours (e.g., Monday, Tuesday, Wednesday from 2pm to 6pm).
And then comes the crucial part: an algorithm that takes this data as input and, considering the estimated effort and required skills for each task, automatically assigns tasks to individuals by their skills, skills level (lower it is, higher the estimated time between the effort range) and their availability. It should provide you with a Gantt chart indicating the estimated completion date and alert you if the estimates go beyond the deadlines.
The good news is that such a tool exists, and it's pretty impressive (I'm not here to promote them, so feel free to PM me if you're interested). However, it does have some usability issues in my opinion, which make it cumbersome to use on a larger scale. Moreover, it doesn't save time on the most important aspect: creating the task list itself! That's the topic we're discussing, after all! 😂
Therefore, the perfect tool should not only handle all the aforementioned logics, but also generate the to-do list itself and assign estimated effort and required skills to the tasks, and let the AI do the rest 😎.
I've also come across a tool that generates fantastic interactive mind maps starting from high level project descriptions. I tried integrating it with the aforementioned tool, but unfortunately this tools doesn't have an API (it's in their roadmap they say).
I even attempted to implement something using OpenAI, creating intriguing prompts that generated cool to-do lists. However, I never had the time to integrate it with the first tool.
Ideally, with the output from that approach and the logic described earlier, you could create a highly detailed plan in less than half an hour and be ready to start coding instead of struggling with plans. Once you have the comprehensive plan in place and a tool that adjusts to changes by providing new estimated completion dates, maintaining and refining the plan becomes completely stress-free.
In conclusion, the perfect project management tool for the AI era is out there, but its pieces haven't been fully integrated yet 😆
You are correct. Without a plan I either feel aimless or I procrastinate to the point that I never get beyond research. There's a book that really helped me plan out new projects, in particular, learning projects. I made a YouTube video about it awhile back which you can find here if you're interested. https://youtu.be/PMSFebugS9A The book I'm referring to is called Ultralearning by Scott Young.
I liked it.
As Eisenhower famously said - plans are nothing, planning is everything.
Hey there! Great article, it resonated with me on a deep level 😂
You're absolutely right, but I have to admit that I've never been good at planning, especially for a completely new project. Creating a plan for a "simple task" like the one you mentioned is relatively easy to do.
However, when it comes to building something more complex, like a digital menu app, it becomes a different story. There are numerous factors to consider: you have different user roles, which requires authentication, authorization, and so on. You'll need at least one backend and one frontend, if not more. The example you provided in the article quickly becomes just a small piece of a much larger puzzle, and that always intimidates me. I worry that maintaining the plan will consume too much of my time.
Personally, I'm a fan of tools and always on the lookout for the best ones to use. So when I start creating a plan for a project like this, I begin jotting down ideas. But in the end, I always find myself thinking, "This takes too much time. How amazing would it be if there were a tool, especially in this era of AI, that could handle planning for me?"
I should clarify that by planning, I don't mean simply creating a to-do list. That's relatively easy nowadays with chat GPT 😉.
By planning, I mean being able to answer a fundamental question: How long will it take to build this?
Believe me, getting an answer to this question is like finding the Holy Grail 😂
To obtain an answer, you need to know the number of people involved, their availability (are they also working on another project?), the order in which tasks should be done, and who should do them. Some tasks may be dependent on others, and so on. Planning involves numerous factors and requires a significant amount of time to maintain.
In this AI era, I believe such time-consuming tasks should no longer be a necessity.
Trust me, I've spent countless hours trying to find something useful, but my searches have never led me to anything meaningful or, at the very least, something that works the way I envision it.
Here's an approach I think could work:
Tasks should be represented as a nested list, where each task can have subtasks. Dependencies and priorities can be defined for tasks. Each task should also have an estimated effort associated with it (expressed as a range, e.g., 8-12 hours) and the required skill set (or even a list of skills). Additionally, individuals should be defined, with assigned skills and skill levels (ranging from 1 to 10 to represent junior to senior), along with their available working hours (e.g., Monday, Tuesday, Wednesday from 2pm to 6pm).
And then comes the crucial part: an algorithm that takes this data as input and, considering the estimated effort and required skills for each task, automatically assigns tasks to individuals by their skills, skills level (lower it is, higher the estimated time between the effort range) and their availability. It should provide you with a Gantt chart indicating the estimated completion date and alert you if the estimates go beyond the deadlines.
The good news is that such a tool exists, and it's pretty impressive (I'm not here to promote them, so feel free to PM me if you're interested). However, it does have some usability issues in my opinion, which make it cumbersome to use on a larger scale. Moreover, it doesn't save time on the most important aspect: creating the task list itself! That's the topic we're discussing, after all! 😂
Therefore, the perfect tool should not only handle all the aforementioned logics, but also generate the to-do list itself and assign estimated effort and required skills to the tasks, and let the AI do the rest 😎.
I've also come across a tool that generates fantastic interactive mind maps starting from high level project descriptions. I tried integrating it with the aforementioned tool, but unfortunately this tools doesn't have an API (it's in their roadmap they say).
I even attempted to implement something using OpenAI, creating intriguing prompts that generated cool to-do lists. However, I never had the time to integrate it with the first tool.
Ideally, with the output from that approach and the logic described earlier, you could create a highly detailed plan in less than half an hour and be ready to start coding instead of struggling with plans. Once you have the comprehensive plan in place and a tool that adjusts to changes by providing new estimated completion dates, maintaining and refining the plan becomes completely stress-free.
In conclusion, the perfect project management tool for the AI era is out there, but its pieces haven't been fully integrated yet 😆
You are correct. Without a plan I either feel aimless or I procrastinate to the point that I never get beyond research. There's a book that really helped me plan out new projects, in particular, learning projects. I made a YouTube video about it awhile back which you can find here if you're interested. https://youtu.be/PMSFebugS9A The book I'm referring to is called Ultralearning by Scott Young.