Our mission.
Film cameras were built to last lifetimes, but the expertise to repair them wasn't. As the original technicians retire, what they know is quietly disappearing, and with it the future of these cameras. Our mission is to stop that. Working alongside technicians, we've built a knowledge base of over 150,000 pages of service data across 896 models, and made it something you can actually ask.
We were as wary of AI hallucination as anyone. What changed our minds was testing it against real service specifications across hundreds of models.
What we found
We tested it hard, expecting to catch it out. Instead, backed by extensive service expertise, it performed far beyond what we expected, particularly at diagnosing faults on cameras whose service data we don't hold.
Cameras of an era share mechanisms: shutters, meters, wind-on, mirror cages. With real specifications to measure against, the model can often work a fault from what it knows about the mechanism family. When it's doing that, it tells you. You're never left guessing whether an answer comes from the data for the camera in front of you or from reasoning across related bodies.
Grounded in known methods
A repair answer that sounds authoritative but isn't is worse than no answer at all. You've already got the top plate off. So the service expertise is trained into the model itself, and every answer is checked back against official specifications and proven repair methods.
It isn't boxed in either. The model will search the web for helpful information, known fault patterns, community write-ups, anything relevant to your repair. But it always reverts back to known methods, filtering out the wrong turns and bodge advice before it reaches you.