Many clients try an off-the-shelf model first. It handles the simple cases and fails on the ones that matter. For those, you need engineers who write recognition algorithms in Python, collect and label the right data, and train a model for your scene. That is what we do, alongside practical LLM features for everyday business software.
Recognition and measurement for your specific objects, conditions and cameras, from data collection to a trained model.
Phone-based scanning pipelines that turn images into measurable 3D models.
Chat, drafting, search over your documents and form filling, with human review where mistakes are costly.
A short proof of concept on your own images, with an accuracy report, before anyone commits to a full build.
We report accuracy on a test set drawn from your real scenes, not on a demo.
We tell you what the proof of concept achieved and what it would take to go further. Then you decide.
Where an error is expensive, the system flags uncertain results for a person to confirm.
A 3D scanning technology company
An end-to-end chain from phone capture to finished 3D model: iOS scanning in Swift, a reconstruction engine in C++, and Python processing stages. Eight repositories, maintained from 2024 to 2026.
Swift · C++ · Python
Full write-up (Chinese)
A US foot-care and orthotics company
Two native Swift apps that capture the foot in 3D and turn the scan into a custom orthotics order. Released on the US App Store, with the first app maintained and updated since 2022.
Swift · 3D scanning SDK · Firebase
Full write-up (Chinese)
An overseas social video start-up
A native Swift app, from version 1.0 in 2024 through a full 2.0 rebuild, together with a Python model service for image-to-text, video and music generation. Released on the US and UK App Stores in 2026 and still being updated.
Swift · Firebase · Python
Our own product
Take a photo, identify the aquarium species, and land on its encyclopedia entry and community posts. Recognition, content and admin share one data model.
Image recognition · Mobile
Full write-up (Chinese)
An AI application company
Three front-ends for AI products (conversational, form-driven and creative) on web and cross-platform mobile, maintained from 2024 to 2026.
Vue · uni-app · LLM APIs
Full write-up (Chinese)That is the most common way clients come to us. Send us a sample of the images where it fails. A proof of concept on your own data tells both of us whether a custom model can close the gap.
It depends on how varied your scene is. We start with what you have, measure, and tell you whether more data or labelling will help.
We choose per project, based on accuracy, cost and where your data is allowed to go, including self-hosted open models.
What the product does, which platforms or devices are involved, and your timing. We reply within three working days, in English, with questions or a short written proposal.
