Thesis: New technologies often benefit some occupations (e.g., robot technicians), while simultaneously harming others (e.g., cashiers). In principle, their negative impact could be mitigated if displaced workers reallocate to expanding occupations. However, lack of skill transferability between occupations often constrains worker mobility, leading to prolonged unemployed. In this dissertation, I study how skill frictions shapes the speed of worker reallocation across occupations. My key contribution is to represent the structure of skill frictions as a network of occupations, where connections represent feasible transitions based on skill similarity. This network perspective uncovers the crucial role of specific 'bridge' occupations in shaping the speed of reallocation. First, I construct the occupation network using expert data on skills and document it is divided in clusters of similar occupations, with few 'bridge occupations' connecting distinct clusters (logistics technician or sales assistant). Leveraging French administrative data, I show that workers transitioning through these ’bridges’ move to distant occupations with higher wages and lower unemployment. Second, I develop a new theory of job search in networked labor markets. My main finding is that job-finding rates in bridge occupations —indicating their local accessibility— have a significant impact on overall reallocation speed. Intuitively, these bridge occupations act as hubs, facilitating transitions between clusters. Moreover, I find that slow transition can lead to important welfare losses. Third, then augment the model with quantitative extensions, leveraging hat-algebra methods to solve counterfactuals without having to estimate large numbers of parameters. Calibrated to French data, the model predicts that robot adoption induces slow reallocation, around 40 quarters, and that this sluggish reallocation reduces welfare gains by approximately 40%— an order of magnitude higher than previous estimates. Moreover, policies targeting bridge occupations can speed-up reallocation, and much more so than policies targeting tight occupations directly. These findings highlight the crucial role of the occupation network in shaping reallocation dynamics and provide new insights for the design of labor market policies.