It seems like an era of doing things this way has arrived.
Traditionally, as engineers, practical skills and knowledge aimed at full-stack capabilities—directly handling coding/infrastructure and such—were important.
Now it seems that the important work is instructing AIs in parallel, receiving results, and evaluating them.
Somehow I ended up subscribing to all 4 AIs, and I set up n8n workflows to automate tasks, while planning the next tasks with other AIs—the switching does seem to age my brain a bit (people really do become the bottleneck).
Vibe coding is now a thing of the past, and it seems like this is the year to master parallel agent engineering.
It looks like this year's goal is becoming how to structure your work and tasks for AI in an organized, automated, and verifiable way. What do you all think~?
