Trust is recognised as a significant and valuable component of socio-technical systems, facilitating numerous important benefits. Many trust models have been created throughout various streams of literature, describing trust for different stakeholders in different contexts. However, when designing a system with multiple stakeholders in their multiple contexts, how does one decide which trust model(s) to apply? And furthermore, how does one go from selecting a model or models to translating those into design? We review and analyse two prominent trust models, and apply them to the design of a trustworthy socio-technical system, namely virtual research environments. We show that a singular model cannot easily be imported and directly implemented into the design of such a system. We introduce the concept of alchemy as the most apt characterization of a successful design process, illustrating the need for designers to engage with the richness of the trust landscape and creatively experiment with components from multiple models to create the perfect blend for their context. We provide a demonstrative case study illustrating the process through which designers of socio-technical systems can become alchemists of trust.