Epitaxy, or growth of a thin crystalline layer upon a crystalline substrate, is a mainstay of both scientific research and modern electronics technology. Strained layer epitaxy of semiconductors can modify band structure, change growth mode, and nucleate dislocations to relieve strain. In this talk, I will discuss recent work from my group on the growth of III-V semiconductors on silicon substrates for application in solar energy and integrated photonics. We employ molecular beam epitaxy (MBE), which has been described as “atomic spray painting” in an ultra-high vacuum environment, for atomically precise growth of high-purity semiconductor layers. MBE was pioneered by Grainger Engineering Hall of Famer, Alfred Cho (BS 1960, MS 1961, PhD 1968) at Bell Laboratories more than 50 years ago. I will first describe a surprising result where n-type doping of strained GaP grown on Si exhibits dislocation densities 30x higher than p-type and undoped layers; dislocations in semiconductors act as non-radiative recombination centers and can greatly harm device efficiencies if left unchecked. I will show that this escalation is not the result of solute drag and therefore likely results from Fermi level effects on dislocation velocity. Understanding the impact of doping on dislocation velocity was important to our demonstration of record-efficiency GaAsP/Si tandem solar cells. Next, I will give a brief background on how strain can drive the self-assembled growth of quantum dots, which can serve as an efficient and dislocation-tolerant active region for laser diodes grown on Si. I will conclude with a discussion of our group’s recent work to integrate visible quantum dot lasers on photonic integrated circuits, opening new pathways for quantum information technologies.