Although Fe–S clusters may assemble spontaneously from elemental iron and sulfur in protein-free systems, the potential toxicity of free Fe2 +, Fe3 +, and S2 − ions in aerobic environments underscores the requirement for specialized proteins to oversee the safe assembly of Fe–S clusters in living cells. Prokaryotes first developed multiprotein systems for Fe–S cluster assembly, from which mitochondria later derived their own system and became the main Fe–S cluster suppliers for eukaryotic cells. Early studies in yeast and human mitochondria indicated that Fe–S cluster assembly in eukaryotes is centered around highly conserved Fe–S proteins (human ISCU) that serve as scaffolds upon which new Fe–S clusters are assembled from (i) elemental sulfur, provided by a pyridoxal phosphate-dependent cysteine desulfurase (human NFS1) and its stabilizing-binding partner (human ISD11), and (ii) elemental iron, provided by an iron-binding protein of the frataxin family (human FXN). Further studies revealed that all of these proteins could form stable complexes that could reach molecular masses of megadaltons. However, the protein–protein interaction surfaces, catalytic mechanisms, and overall architecture of these macromolecular machines remained undefined for quite some time. The delay was due to difficulties inherent in reconstituting these very large multiprotein complexes in vitro or isolating them from cells in sufficient quantities to enable biochemical and structural studies. Here, we describe approaches we developed to reconstitute the human Fe–S cluster assembly machinery in Escherichia coli and to define its remarkable architecture.