Extracellular signal-regulated kinase 1/2 (ERK1/2) is a serine/threonine protein kinase in eukaryotic cells and belongs to the mitogen-activated protein kinase (MAPK) family. ERK1/2 is essential for signaling from surface receptors to the nucleus. Activated ERK1/2 phosphorylates substrates in the nucleus or cytoplasm, causing certain proteins to be expressed or activated, regulating cell proliferation, differentiation, death, and other functions. ERK1/2 is abundantly expressed in several forms of ischemia-reperfusion injury (IRI). Caffeic acid (3,4-dihydroxy cinnamic acid), as previously reported, interacted directly with ERK1/2 and reduced its actions in vitro. It is the most abundant phenolic component found in coffee. Moreover, it is reported to have a variety of pharmacologic effects, including anti-inflammatory, immunomodulatory, antioxidant, and anticancer effects. In the present study, we employ a deep learning protocol to generate novel and effective anti-ERK1/2 drugs by modifying the chemical structure of caffeic acid, aiming to improve its inhibition performance. Conventional methods in the drug discovery and development process can take several years, and the cost of bringing a novel drug to market can be around two billion dollars. Instead of all these efforts, computer-aided drug design (CADD) can be an effective, rapid, and cost-efficient method to design novel drugs. In the current study, a molecular docking as well as a molecular dynamics study will be executed to explore the effectiveness of the generated drugs. To get a deeper insight into the potency of the studied inhibitors, drug-likeness analysis will be applied to check the bioavailability of the best-ranked drugs.