Applied-focused courses (like modeling, optimization, etc.) are very broad, and I think that's where some application gets lost. For me, learning proofs within

Applied-focused courses (like modeling, optimization, etc.) are very broad, and I think that's where some application gets lost. For me, learning proofs within an applied-subject-domain (like partial differential equations) helped me a lot, because it was a different "flavor" of proofs, where you could understand that applied domain much much deeper.