Let $f:X\to Y$ be a map between first countable Hausdorff spaces for which $f^{-1}(K)$ is compact, for all compact $K\subseteq Y$. Show that $f$ is a closed map.
My attempt:
Let $F\subseteq X$ be closed in $X$. I want to show that $Y\backslash f(F)$ is open in $Y$. Take $y\in Y\backslash f(F)$, then $\forall z\in f(F)\subseteq Y: z\ne y. $ Since $Y$ is Hausdorff, there exist $U_z, V_z \in \tau_Y$ such that $z\in U_z, y\in V_z, U_z \cap V_z =\emptyset$ (as this holds for all $z\in f(F)$, we will specify this $z$-dependency when writing down the Hausdorff sets). Now, I believe that $$ Y\backslash f(F) = \bigcup_{z\in f(F)} V_z \in \tau_Y.$$ Proof of the equality: Let $y\in Y\backslash f(F)$, then $y\in V_w$ for some $w\in f(F)$ because $Y$ is Hausdorff and $y\ne w$, which proves the $\subseteq$-inclusion. Now let $y$ be an element of the union, then there is some $z\in f(F)$ for which $y\in V_z\subseteq Y$. It remains to show that $y$ cannot be an element of $f(F)$. Suppose it is, then $V_z \cap f(F)\ne \emptyset$. Note that $U_z\cap f(F)$ is an open neighborhood of $z$ (in $f(F)$), for which the intersection with $V_z$ is empty (as $V_z$ and $U_z$ are disjoint, Hausdorff). Now, I'm stuck. I can't find an argument to show that this assumption will result in a contradiction. I haven't used the first countability and the compactness property of $f$ yet, but I can't find a good application for them.
I would like to know if my approach is correct, if so, how to finish the proof?
Thanks.