Instability under non-native processing conditions, especially at elevated temperatures, is a major factor preventing widespread adoption of biocatalysts for industrial synthesis. Therefore, enzymes with higher thermostability are necessary to create economically viable biocatalytic processes. Here we test a simple, broadly applicable method for the generation of thermostable proteins on two distinct enzymes. A crucial distinction of many redox enzymes used to synthesize chiral compounds is the need of cofactors (e.g., NAD(P)(H)) for function. Due to the prohibitively high prices of nicotinamide cofactors and the varied non-native conditions of solvent, pH, and temperature used in the synthesis of many compounds, a robust cofactor regenerating enzyme is required for economical synthesis of fine chemicals via biocatalysis. We aim to increase the temperature stability of glucose dehydrogenase (GDH) via a structure-guided consensus concept. GDH is a homotetramer capable of accepting NAD+ or NADP+ with high specific activity, making it commonly used as a cofactor recycle enzyme, as well as an attractive target for improvement. While semi-rational methods based on sequence alignments (i.e., the consensus concept) are promising, these methods are limited to using highly homologous protein families to identify a reduced number of candidate mutations. In our case twelve GDHs were aligned in addition to the wild-type template GDH from Bacillus subtilis. These sequences ranged in homology from 24.8-61.1% to the template and provided a higher number of amino acid candidates than high homology consensus approaches. The candidates were then sieved using amino acid propensities and structural information (i.e., crystal structure or homology-based model) to select the final residues for substitution. Using this approach we generated 23 variants, 12 of which showed higher thermal stability than the wild-type GDH. The best variant engineered to date has a half-life of 4 days at 65oC. In addition, the three most stabilizing single mutations were transferred to two GDH homologs from Bacillus thuringiensis and Bacillus licheniformis. The thermal stability of the GDH variants was increased as expected. The resulting stability changes provides further support that these residues are critical for stability of GDHs and reinforces the success of the consensus approach for identifying stabilizing mutations. Pencillin G acylase (PGA) is the workhorse in the synthesis of ƒ"-lactam antibiotics as well as precursors such as 6-aminopenicilanic acid. Currently, the worldwide production of penicillins and 6-APA surpass the 30,000 and 8,000 metric tons per year, respectively. A more stable PGA would improve its competence in a commercial process. We used the stucture-guided consensus method to limit the number of PGA variants created. Of our 21 single-mutants, 10 were found experimentally to be more thermostable than the wild-type PGA°Xa significant increase in success rate when compared to other traditional techniques such as directed evolution. We demonstrate the broad applicability of structure-guided consensus method by testing it in two distinct cases successfully. Furthermore, by transferring mutations to other homologs we show that stability of homologs of unknown structure can be improved using this approach.