I trained as a chemical engineer. Two pieces of that work went through outside review and came out the other side: a granted US patent and a peer-reviewed paper. They are why I still think in processes, materials and failure modes.
A granted United States patent for a device that suppresses and mitigates flame propagation. It came out of hands-on process engineering at a US plastics manufacturer, where I was named on the patent.
Research on how surface chemistry governs molecular adhesion at a conducting-polymer interface, from my undergraduate years in chemical engineering at Purdue University. Published in Colloids and Surfaces A: Physicochemical and Engineering Aspects.
A plant runs on steps that exist for reasons nobody wrote down. Software systems do too. I read the old process until I know what is safe to remove.
A hypothesis is not a result. In a lab, and now in production debugging, I measure first and keep a written record of what was tried and why it failed.