So my work is mandating agentic coding. One thing I have found is that it shifts to the right (as opposed to you know...shiftLeft), both in terms of quality and understanding. I have adjusted by scrutinizing the output (code) much deeper than if I owned more of the workflow (discovery at minimum).
But ofc, that annuls much of the time saved using an agentic workflow. #programming #agents #AI
And yes, I did have an agent review the code before I reviewed it. Also had another create an ffbd for me with line annotations. Still ended up having to go deep on that code review at the end to understand it.

Also, this is one workflow at a time. I don't get the talk of parallelization of agents. How on earth do you deal with all the context switching? I think it would kill me tbh.

It takes all of my cognitive bandwidth just reading and understanding the outputs.