You have some answers already, but let me just point out what might be a source of confusion: if a function is defined piecewise, it can be differentiated piecewise, except at the endpoints which must be checked individually.
For instance, your function is
$$
f(x) = |x| = \begin{cases} x & x > 0 \\ -x & x < 0 \\ 0 & x = 0\end{cases}
$$
It's tempting to say, “Well, $x$, $-x$, and $0$ are all differentiable functions, so
$$
\color{red}{
f'(x) = \begin{cases} 1 & x > 0 \\ -1 & x < 0 \\ 0 & x = 0\end{cases}
}
$$
and therefore $f$ is differentiable for all $x$.”
But actually, we can only differentiate on the interior of the pieces. That is,
$$
f'(x) = \begin{cases} 1 & x > 0 \\ -1 & x < 0 \end{cases}
$$
is true. But to check whether $f$ is differentiable at $0$, we need to decide if
$$
\lim_{h\to 0} \frac{f(0+h) - f(0)}{h} = \lim_{h\to 0} \frac{|h|}{h}
$$
exists. In fact, this limit does not exist.
The moral of the story is that differentiability requires checking not just a point, but a neighborhood around that point. This is why when it can fail to be differentiable at the transition points of piecewise-defined functions.