The ongoing climate change has put a spotlight on rapid evolutionary processes that could aid species adapt to new environments. However, many questions remain unanswered: What is the genetic architecture traits influencing fitness across environments? Is this genomic architecture predictable? Can we understand genetic constraints across multiple adaptive traits? How is genetic variation lost during extinction? To address these questions, we combine statistical genomics with experimental ecology and genetic engineering approaches using the plant species Arabidopsis thaliana as our experimental climate change genetics model and scaling up insights with publicly available genomes of diverse plant species.