We built this because we were tired of losing track of what we read.

A PhD student reads roughly 1,500 papers over five years. They remember maybe 30 of them. The rest live in folders with names like Downloads/Mendeley/Unsorted/misc.

We tried every tool. NotebookLM lost context. ChatGPT hallucinated citations. Zotero remembered files but not reasoning. We wanted something that held on to why we cared about a paper, not just that it existed.

SuperLabCoat is the system we wished we had in grad school: papers flow in, claims get linked, experiments support or contradict hypotheses, and when it's time to write, the draft comes out of your actual work, not a language model's imagination.

Your research memory, externalized.