7,8-Diaminopelargonic acid synthase (BioA) of Mycobacterium tuberculosis (Mtb) is a recently validated target for therapeutic intervention in the treatment of tuberculosis (TB). We herein report our fragment based inhibitor design of Mtb BioA. Using differential scanning fluorimetry (DSF) fragment screening, the Maybridge Ro3 library of 1000 molecules was screened. Twenty-one compounds giving rise to Tm shifts exceeding ±2°C were then investigated in crystallographic experiments. Six fragments have been co-crystallized with BioA to characterize binding. Each compound has a unique binding mode, and subtle variations in ligand binding site geometry are induced upon binding of different fragment molecules. Binding affinities of the fragments were characterized via isothermal titration calorimetry (ITC). A fragment extension strategy was used to rationally optimize these fragment hits. A commerce based SAR was used and identified 50 compounds containing the core of one of the fragments. These compounds were further screened virtually and experimently by DSF. Four optimized BioA ligands from fragment optimization were validated by X-ray crystallography, including a potent aryl hydrazine inhibitor of BioA that reversibly modifies the pyridoxal-5′-phosphate (PLP) cofactor. Binding affinities of these ligands have been characterized by ITC or kinetic assay. The six fragment complex structures were also used for optimization of HTS lead compounds. Six HTS lead compounds were co-crystallized with BioA at high resolution. Design of optimized compounds was by overlapping the fragments and HTS lead binding conformations in the BioA active site. Molecules predicted to have better potency were proposed. Two N-aryl piperazine inhibitors of BioA from HTS optimization were characterized using X-ray crystallography and ITC. One inhibitor that combines features of one HTS lead and one fragment was confirmed with improved binding affinity by ITC.