To establish that two vector spaces are isomorphic, we must find an isomorphism between them, an invertible linear transformation (Definition IVS). The kernel of the linear transformation $T$, $\krn{T}$, is a subspace of $U$, while the null space of the matrix representation, $\nsp{\matrixrep{T}{B}{C}}$ is a subspace of $\complex{n}$. The function $\vectrepname{B}$ is defined as a function from $U$ to $\complex{n}$, but we can just as well employ the definition of $\vectrepname{B}$ as a function from $\krn{T}$ to $\nsp{\matrixrep{T}{B}{C}}$.
We must first insure that if we choose an input for $\vectrepname{B}$ from $\krn{T}$ that then the output will be an element of $\nsp{\matrixrep{T}{B}{C}}$. So suppose that $\vect{u}\in\krn{T}$. Then \begin{align*} \matrixrep{T}{B}{C}\vectrep{B}{\vect{u}} &=\vectrep{C}{\lt{T}{\vect{u}}}&&\knowl{./knowls/theorem.FTMR.knowl}{\text{Theorem FTMR}}\\ &=\vectrep{C}{\zerovector}&&\knowl{./knowls/definition.KLT.knowl}{\text{Definition KLT}}\\ &=\zerovector&&\knowl{./knowls/theorem.LTTZZ.knowl}{\text{Theorem LTTZZ}}\\ \end{align*}
This says that $\vectrep{B}{\vect{u}}\in\nsp{\matrixrep{T}{B}{C}}$, as desired.
The restriction in the size of the domain and codomain $\vectrepname{B}$ will not affect the fact that $\vectrepname{B}$ is a linear transformation (Theorem VRLT), nor will it affect the fact that $\vectrepname{B}$ is injective (Theorem VRI). Something must be done though to verify that $\vectrepname{B}$ is surjective. To this end, appeal to the definition of surjective (Definition SLT), and suppose that we have an element of the codomain, $\vect{x}\in\nsp{\matrixrep{T}{B}{C}}\subseteq\complex{n}$ and we wish to find an element of the domain with $\vect{x}$ as its image. We now show that the desired element of the domain is $\vect{u}=\vectrepinv{B}{\vect{x}}$. First, verify that $\vect{u}\in\krn{T}$, \begin{align*} \lt{T}{\vect{u}} &=\lt{T}{\vectrepinv{B}{\vect{x}}}\\ &=\vectrepinv{C}{\matrixrep{T}{B}{C}\left(\vectrep{B}{\vectrepinv{B}{\vect{x}}}\right)}&&\knowl{./knowls/theorem.FTMR.knowl}{\text{Theorem FTMR}}\\ &=\vectrepinv{C}{\matrixrep{T}{B}{C}\left(\lt{I_{\complex{n}}}{\vect{x}}\right)}&&\knowl{./knowls/definition.IVLT.knowl}{\text{Definition IVLT}}\\ &=\vectrepinv{C}{\matrixrep{T}{B}{C}\vect{x}}&&\knowl{./knowls/definition.IDLT.knowl}{\text{Definition IDLT}}\\ &=\vectrepinv{C}{\zerovector_{\complex{n}}}&&\knowl{./knowls/definition.KLT.knowl}{\text{Definition KLT}}\\ &=\zerovector_{V}&&\knowl{./knowls/theorem.LTTZZ.knowl}{\text{Theorem LTTZZ}}\\ \end{align*} Second, verify that the proposed isomorphism, $\vectrepname{B}$, takes $\vect{u}$ to $\vect{x}$, \begin{align*} \vectrep{B}{\vect{u}}&=\vectrep{B}{\vectrepinv{B}{\vect{x}}}&&\text{Substitution}\\ &=\lt{I_{\complex{n}}}{\vect{x}}&&\knowl{./knowls/definition.IVLT.knowl}{\text{Definition IVLT}}\\ &=\vect{x}&&\knowl{./knowls/definition.IDLT.knowl}{\text{Definition IDLT}} \end{align*}
With $\vectrepname{B}$ demonstrated to be an injective and surjective linear transformation from $\krn{T}$ to $\nsp{\matrixrep{T}{B}{C}}$, Theorem ILTIS tells us $\vectrepname{B}$ is invertible, and so by Definition IVS, we say $\krn{T}$ and $\nsp{\matrixrep{T}{B}{C}}$ are isomorphic.