transforming ordinary generating function into exponential generating function
There may be some cases where complex variables, the residue theorem and the residue at infinity are helpful. Suppose your OGF is $f(z)$ and the desired EGF is $g(w).$ Then we have
$$g(w) = \sum_{n\ge 0} \frac{w^n}{n!} \frac{1}{2\pi i} \int_{|z|=\epsilon} \frac{1}{z^{n+1}} f(z) \; dz.$$
This will simplify together with some conditions on convergence to give $$g(w) = \frac{1}{2\pi i} \int_{|z|=\epsilon} \frac{f(z)}{z} \sum_{n\ge 0} \frac{1}{n!} \frac{w^n}{z^n} \; dz \\ = \frac{1}{2\pi i} \int_{|z|=\epsilon} \frac{f(z)}{z} \exp(w/z) \; dz.$$
Example I. Suppose $$f(z) = \frac{1}{1-z},$$ which yields $$\frac{1}{2\pi i} \int_{|z|=\epsilon} \frac{1}{1-z} \frac{1}{z} \exp(w/z) \; dz.$$
Now for $z=R\exp(i\theta)$ with $R$ going to infinity we have $$2\pi R\times \frac{1}{R^2} \times \exp(|w|/R) \rightarrow 0$$ as $R\rightarrow \infty$ so this integral is $$- \mathrm{Res}_{z=1} \frac{1}{1-z} \frac{1}{z} \exp(w/z)$$ and we get $$g(w) = \exp(w)$$ which is the correct answer.
Example II. This time suppose that $$f(z) = \frac{z}{(1-z)^2}$$ so that we should get $$g(w) = \sum_{n\ge 1} n \frac{w^n}{n!} = w \sum_{n\ge 1} \frac{w^{n-1}}{(n-1)!} = w\exp(w).$$
The integral formula yields $$\frac{1}{2\pi i} \int_{|z|=\epsilon} \frac{z}{(1-z)^2} \frac{1}{z} \exp(w/z) \; dz \\ = \frac{1}{2\pi i} \int_{|z|=\epsilon} \frac{1}{(1-z)^2} \exp(w/z) \; dz.$$
The residue at infinity is zero as before and we have $$\exp(w/z) = \sum_{n\ge 0} \left. (\exp(w/z))^{(n)}\right|_{z=1} \frac{(z-1)^n}{n!}$$
The coefficient on $(z-1)$ is $$\left. -\frac{1}{z^2} w\exp(w/z)\right|_{z=1} = - w\exp(w)$$
which is the correct answer taking into account the sign flip due to $z=1$ not being inside the contour.
Remark. Good news. The sum in the integral converges everywhere.
Addendum: somewhat more involved example. The OGF of Stirling numbers of the second kind for set partitions into $k$ non-empty sets is
$$\sum_{n\ge 0} {n\brace k} z^n = \prod_{q=1}^k \frac{z}{1-qz}.$$
We thus have that $$g(w) = \frac{1}{2\pi i} \int_{|z|=\epsilon} \frac{1}{z} \exp(w/z) \prod_{q=1}^k \frac{z}{1-qz} \; dz \\ = \frac{(-1)^k}{2\pi i} \int_{|z|=\epsilon} \frac{1}{z} \exp(w/z) \prod_{q=1}^k \frac{z}{qz-1} \; dz \\ = \frac{(-1)^k}{k! \times 2\pi i} \int_{|z|=\epsilon} \frac{1}{z} \exp(w/z) \prod_{q=1}^k \frac{z}{z-1/q} \; dz.$$
Computing the sum of the residues at the finite poles not including zero we get $$\frac{(-1)^k}{k!} \sum_{q=1}^k q\exp(qw) \times \frac{1}{q} \prod_{m=1}^{q-1} \frac{1/q}{1/q-1/m} \prod_{m=q+1}^k \frac{1/q}{1/q-1/m} \\ = \frac{(-1)^k}{k!} \sum_{q=1}^k \exp(qw) \prod_{m=1}^{q-1} \frac{m}{m-q} \prod_{m=q+1}^k \frac{m}{m-q} \\ = \frac{(-1)^k}{k!} \sum_{q=1}^k \exp(qw) \frac{k!}{q} \prod_{m=1}^{q-1} \frac{1}{m-q} \prod_{m=q+1}^k \frac{1}{m-q} \\ = \frac{(-1)^k}{k!} \sum_{q=1}^k \exp(qw) \frac{k!}{q} \frac{(-1)^{q-1}}{(q-1)!} \frac{1}{(k-q)!} \\ = - \frac{1}{k!} \sum_{q=1}^k \exp(qw) (-1)^{k-q} {k\choose q} \\ = -\left(\frac{(\exp(w)-1)^k}{k!} - \frac{(-1)^k}{k!}\right).$$
This is a case where the residue at infinity is not zero. We have the formula for the residue at infinity $$\mathrm{Res}_{z=\infty} h(z) = \mathrm{Res}_{z=0} \left[-\frac{1}{z^2} h\left(\frac{1}{z}\right)\right]$$
This yields for the present case $$- \mathrm{Res}_{z=0} \frac{1}{z^2} z \exp(wz) \prod_{q=1}^k \frac{1/z}{1-q/z} = - \mathrm{Res}_{z=0} \frac{1}{z} \exp(wz) \prod_{q=1}^k \frac{1}{z-q} \\ = - \frac{1}{k!} \mathrm{Res}_{z=0} \frac{1}{z} \exp(wz) \prod_{q=1}^k \frac{1}{z/q-1} \\ = - \frac{(-1)^k}{k!} \mathrm{Res}_{z=0} \frac{1}{z} \exp(wz) \prod_{q=1}^k \frac{1}{1-z/q} = - \frac{(-1)^k}{k!}.$$
Adding the residue at infinity to the residues from the poles at $z=1/q$ we finally obtain $$-\left(\frac{(\exp(w)-1)^k}{k!} - \frac{(-1)^k}{k!}\right) - \frac{(-1)^k}{k!} = - \frac{(\exp(w)-1)^k}{k!}.$$
Taking into account the sign flip we have indeed computed the EGF of the Stirling numbers of the second kind $$\sum_{n\ge 0} {n\brace k} \frac{z^n}{n!}$$ as can be seen from the combinatorial class equation $$\def\textsc#1{\dosc#1\csod} \def\dosc#1#2\csod{{\rm #1{\small #2}}}\textsc{SET}(\mathcal{U}\times\textsc{SET}_{\ge 1}(\mathcal{Z}))$$ which gives the bivariate generating function $$G(z, u) = \exp(u(\exp(z)-1)).$$