Our Story

The Supply Engine was born out of necessity. After leaving medical school to work with technology, and with a new child, our founder started an Amazon business as a transition project. His background in scientific research taught him how often data analysis was done by hand. He quickly learned that product sourcing was no different. Discovering products to sell, and finding good suppliers, took hours of manual work.

The underlying problem was the same. Tools could interpret structured information – things like prices or sales volumes with numbers and clear patterns. But sourcing usually involves unstructured information – tasks like analyzing product titles, searching for reputable suppliers, interpreting text, and making a decision using scattered information. Most of this work had to be done by hand, so our founder began testing ways to automate this research.

While an early version of the Supply Engine worked for internal use, several barriers stood between this prototype and a viable product. Nevertheless, when our founder demonstrated it to a colleague, their enthusiasm revealed this was a promising direction. "I didn't think it was possible, but seeing this is eye-opening. I can't stop telling my wife about it."

After a further year of development, timing proved important. With language models becoming powerful, affordable, and well enough understood to overcome the few remaining barriers, it is finally becoming possible to create a reliable sourcing engine.

We believe the Supply Engine represents more than task automation. The ability to reliably interpret unstructured data means human judgment can be automated at scale. Our vision is to expand beyond product sourcing into a broader platform for online businesses, saving countless hours by transforming how they perform unstructured research.