This article investigates the asymmetric relationship between the algorithmic control of digital platforms and the agency of ride-hailing drivers, arguing that worker responses are not homogeneous but are instead organized along a continuum of differentiated agencies. Through a qualitative study based in Mexico City, five driver archetypes were identified, whose micro-strategies were analysed through the frameworks of structural vulnerability and strategic power. The main theoretical contribution of this study is the conceptualization of differentiated precarity, which posits that the vulnerability inherent in contractual instability is moderated by the driver's social trajectory and prior economic capital. Through the study it is demonstrated that profiles with greater capital (i.e., the Strategic, the Artisan, the Precarious in Transit) employ micro-strategies of arbitrage and shadow organizing to fragment algorithmic control. In contrast, those with critical economic dependence (i.e., the Algorithmic, the Excluded) resort to tactical submission and forced adaptation as survival strategies. The study concludes that autonomy in digital work is reconfigured as a struggle to insulate one's identity and resources from the system's total control. The resulting policy implications suggest the need for differentiated regulatory solutions that protect the most vulnerable segment. The study's limitations, including social desirability bias and its specific geographical focus, pave the way for future longitudinal research and the development of mixed-methods approaches to quantify the economic impact of collective agency.