Computer vision platform for barcode, text, and ID capture across mobile and fixed devices
Scandit deploys computer vision (PyTorch, TensorFlow, C++) to enable data capture from barcodes, text, IDs, and objects across smartphones, handheld computers, drones, and fixed cameras. The stack—anchored in deep learning and mobile SDKs (iOS, Android, React Native, Flutter)—reflects a mature ML-first architecture. Current hiring acceleration spans engineering, marketing, and support, while active projects reveal internal focus on semantic document parsing and workflow automation, suggesting both product expansion and operational scaling challenges.
Scandit builds a smart data capture platform used by enterprise teams in retail, logistics, healthcare, and manufacturing to automate inventory, identity verification, and process workflows. The product captures and interprets physical data (barcodes, printed text, document IDs, object features) using on-device computer vision, with speed and accuracy as core differentiators. The company operates as a SaaS platform with direct enterprise sales, implementation support, and customer success services embedded in the go-to-market model. Scandit is privately held, based in Zurich, and serves mid-market to large enterprises globally.
Scandit's core stack includes PyTorch, TensorFlow, and C++ for model training and inference, with JAX and Transformers for advanced feature extraction. The platform runs on iOS, Android, React Native, and Flutter for mobile deployment, plus native C++17 for performance-critical paths.
Active projects include semantic document parsing for customer use cases, workflow automation roadmap, digital twins for document identification, and mobile app experience design. The team is also addressing internal pain points around manual ticket handling, ticket visibility, and Snyk vulnerability reduction.
Other companies in the same industry, closest in size