Day 55 posts: LinkedIn and X
The angle is the discipline of designing before coding. Swap in your own voice.
Day 55 of 60 days of system design. The capstone begins, and day one has no code.
I am building one system over the next four days: design it, build one real slice, break that slice under load, write it up. And today, on purpose, I wrote the design document first. No editor, no typing, just words.
This felt slow and slightly silly. It is the most useful habit I have picked up in this whole course.
Writing the design first does three things. It forces scope: I wrote down the three things I am building and, more importantly, the things I am not building, so I ship one real feature instead of ten half ones. It surfaces the hard decisions while they are still cheap words, not expensive code. And it gives me something to check reality against on day 57, when I load test the thing and find out whether my design held.
The rule I set for picking the slice: choose the one where a measured number will surprise me. This whole course has been predict a number, measure it, learn from the gap. So my capstone slice is one where I can predict throughput, build it, push it until it breaks, and find out exactly where I was wrong.
Anyone can start typing. Deciding what you will build, and what you will not, before you build it, is the thing that actually ships. That outlasts any interview.
Tomorrow: the build.
Code and notes: github.com/anurag629/system-design-60-days
#systemdesign #softwareengineering #learninginpublic
X thread
1/
Day 55 of 60 days of system design. The capstone begins, and day one has zero code.
One system, four days: design it, build a slice, break it under load, write it up.
Today I wrote the design doc first. Just words.
2/
Felt slow and silly. It is the most useful habit in the whole course.
Writing the design first forces scope: the three things I’m building, and the things I’m NOT. Ship one real feature, not ten half ones.
3/
It also surfaces the hard decisions while they’re cheap words, not expensive code. And it gives me something to check against when I load test on day 57.
4/
My rule for picking the slice: choose the one where a measured number will surprise me.
Predict throughput, build it, push until it breaks, learn from the gap. Same as every day of this course.
5/
Anyone can start typing. Deciding what you’ll build, and what you won’t, before you build it, is what ships.
Tomorrow: the build.
Code: github.com/anurag629/system-design-60-days