Hematopoietic stem cells (HSCs) are responsible for the continuous production of all blood cells. This unique ability has made it possible to successfully use HSCs in the clinical setting to remedy various blood disorders. However, despite six decades of research, the full potential of HSCs to treat patients with hematological malignancies is still restricted by the low number of functional stem cells that can be isolated from different sources. In vitro HSC expansion has been widely considered a potential solution but has proven to be extremely difficult due to the rapid loss of functional potential of cells in the culture outside of their natural microenviron-mental niche. In order to achieve robust HSC expansion in vitro, it is crucial to better under-stand the mechanisms that regulate HSC fate choices. Moreover, the absence of reliable mark-ers for identifying functional HSCs on a culture dish necessitates the use of expensive and time-consuming in vivo assays. The discovery of predictive phenotypic markers for HSC potential could address this problem. Many platforms for analyzing HSC fate at the single-cell level have been reported, but they generally suffer from poor cell viability, limited throughput, unreliable phenotyping by immunocytochemistry, and the difficulty of performing long-term cell cultures. Additionally, these platforms are usually fabricated from polydimethylsiloxane (PDMS), a poly-mer known to have several disadvantages for cell-based applications. This thesis addresses these shortcomings through the development of a robust microchip sys-tem for the high-resolution, live-cell analysis of hundreds to thousands of single stem cells and their daughter cells. The platform combines key advantages of microfluidic technology (for cell lineage analysis) and hydrogel microwell array technology (for gentle cell capture and long-term culture). The platform is fabricated from two different layers of hydrogel to create a closed ar-ray of grooves, where cells’ movements are restricted and grow along one axis. This design en-ables an unambiguous tracking of a single cell and all of its progeny over several generations. These two layers are produced from different biocompatible hydrogels: poly (2-hydroxyethyl methylacrylate) / trimethylolpropane trimethacrylate (pHEMA-TMPTMA), a copolymer acting as a topographically structured cellular substrate that can be closed with a polyethylene glycol (PEG) hydrogel lid. These two parts are combined together and covalently bonded to each oth-er after the seeding of live cells into the grooves of the bottom part. To assess the potential of the platform for single stem-cell analyses, HSCs were cultured and their behavior (including cell fate, proliferation kinetics, and functionality) were thoroughly characterized. To analyze imaging data obtained from single-cell time-lapse experiments performed on the platform, an algorithm was developed to automatically identify and track single HSCs and their daughter cells, efficiently extracting various parameters, such as single-cell proliferation kinet-ics, cell fate choices, and lineage relationships. Mitochondria have recently been identified as key modulators of HSC function; however, their specific role in fate regulation has remained obscure. Therefore, the single-cell analysis plat-form was applied to measure the distribution of mitochondrial mass in dividing HSCs and their progeny to assess a potential correlation with HSC fate choices. This