Caenorhabditis elegans is a performant model system for studying human biological processes and diseases and for pre-clinical phenotyping screenings of compounds. Microfluidics has been instrumental in enabling C. elegans-based drug assays, facilitating the worm handling and bringing robust and reproducible analysis protocols. In this Thesis, I present phenotypic assays on C. elegans, enabled by microfluidic devices conveniently designed. First, I propose a strategy to track different phenotypic traits of individual worms over their life cycle. Phenotypes - as duration of the different developmental stages, length, diameter and fluorescent stress response at each developmental stage, fertility and sex fate - were quantified for each animal, resulting in high-content phenome data. The method was validated by analyzing the worm response to pharmacological and genetic compounds known to activate mitochondrial stress-response pathways in different species. Interestingly, the analysis revealed specific sub-populations that allowed separating single worms as responders or non-responders to a treatment, thereby elucidating pharmaceutical or therapeutic responses still overlooked. Second, I propose a design-of-experiment approach to characterize the concentration-dependent effect of the drug doxycycline on the duration of C. elegans development. 13 experiments were performed following a Doehlert design, where different doxycycline concentrations were tested, while varying also temperature and food amount, which are known to influence the duration of the C. elegans development. A microfluidic platform was designed to test the doxycycline effect on isolated worms, with full control of temperature and feeding over the development. Our approach allowed maximizing the understanding of the effect of this antibiotic on the C. elegans development and paving the way towards a standardized and optimized drug testing process. Third, envisioning a platform where C. elegans on-chip feeding is fully controlled and food consumption could be quantified as additional phenotype, I describe a method to measure the concentration of solutions of Escherichia coli bacteria. I designed an opto-electronic device that was integrated with and optically aligned to a microfluidic chip. The absorbance of a bacterial solution was calculated according to the Lambert-Beer’s law. The device was tested and calibrated, obtaining a linear relation between the absorbance and the concentration of the solutions. Afterwards, the change of the turbidity of the solution on the long-term was assessed. Based on our observations, a simple mathematical model for the change of absorbance over time was hypothesized. In another study I performed a motility-based phenotypic assay on C. elegans, to quantify the worm glucose uptake. Adult worms were isolated in a micro-chamber, a glucose solution was injected and short videos were recorded, from which a motility score was extracted. Last but not least, in the frame of a collaborative project, I designed a 3D-printed micro-channel to immobilize a single adult worm to perform metabolic phenotyping on a worm subsection via nuclear magnetic resonance spectroscopy. Envisioning a pivotal role for C. elegans in the drug screening pipeline, a unique microfluidic platform, integrating all the phenotypic assays described in this Thesis, would allow for robust, fast and cost-effective high-content screenings of compounds on this simple yet powerful model.