I have stumbled upon an interesting linear operator. I am not sure if it is compact: $$ T: L^2([0, 1]) \rightarrow L^2([0, 1]), \quad (Tf)(x) := \int^1_0 (x-y)^2 f(y)~\mathrm{d}y. $$
I tried to use the following argument: Let $(f_k)_{k \in \mathbb{N}} \subseteq L^2([0, 1])$ be bounded and let $(f_{n_k})_{k \in \mathbb{N}} \subseteq L^2([0, 1])$ be a subsequence with weak limit $f \in L^2([0, 1])$. Then: $$ \lVert Tf_{n_k} - Tf \rVert_{L^2([0, 1])}^2 = \int_0^1 \left \lvert \int^1_0 (x-y)^2 (f_{n_k}(y) - f(y))~\mathrm{d}y \right \rvert^2 ~\mathrm{d}x \leq \sup_{x \in [0, 1]} \left \lvert \int^1_0 (x-y)^2 (f_{n_k}(y) - f(y))~\mathrm{d}y \right \rvert^2 $$ I can still prove that this $\sup$ is attained on some $x_{n_k} \in [0, 1]$ but this does not help me because of the dependancy on $n_k$. Otherwise I would have used weak convergence at that point.
Is there an easier way or this is operator even compact?
It's a finite rank operator and thus compact: $$(Tf)(x)=A\,x^2-B\,x+C,$$ where $A=\int^1_0f(y)\,dy, B=2\int^1_0y\,f(y)\,dy, C=\int^1_0y^2\,f(y)\,dy$, so the image of a bounded set is a bounded subset of a 3-dimensional subspace of $L^2([0, 1])$.