Everyone is bad at optimizing their code that people are cooking up wild conspiracy theories about Chinese companies smuggling hundreds of millions of dollars of GPUs into the country…
rather than admit that our whole industry has skill issues.
I find that this is because at your day job in big tech, you are constantly blocked by organizational mis-alignments.
The bigger the company, the more unique plans and goals an organization has. This causes friction at every step. It slows everyone down.
This is also why I
We need a counter-culture design movement that appreciates products that are difficult to use initially, but tremendously fast, easy, accurate, explicit and efficient after topping the learning curve.
Ease-of-use is oversold. It's making everything suboptimal and slow.
I think it’s because if you’re someone who doesn’t know a lot about a domain, LLMs look incredible.
If you’re like Linus, and an industry leading expert, you can look at the output and spot every single mistake being made.
Wages aren’t the figure to look at when it comes to pay arbitrage. It’s “cost of employment”
The $30k gap between US and UK/EU is almost 100% payroll and pension taxes that the company has to pay for the employees
Companies spend about the same to employ someone in both places
Something about the latest layoffs in big tech feels different. Tech used to have a policy of “keeping talent”. Even when big cuts were made, core people were kept and shuffled
The latest Microsoft cuts were comparatively small and a lot of core people with huge value were cut
Disagree.
It violates the Single Responsibility Principle. createfile() should just create a file, not manage a filesystem hierarchy... and having it do so leads to unexpected side effects, like the creation of folders that were not already present.
in RTS games like age of empires
at the beginning you are focused on the micro - acquiring basic resources takes up 100% of your brain
over time you build up enough systems so that a lot of this can happen without needing much from you
by the end of the game you're hardly
One of the more interesting parts of software engineering is debugging something in an environment where you cant directly observe what’s happening, and need to craft experiments to infer the internal state of a system
It’s always frustrating, but so satisfying when you crack it