Tech Stack

Description

You will work on creating digital twins of complex labels and identification documents, evolving our product by adding features and optimizing existing offerings using computer vision and deep learning to improve information extraction from printed documents.

Requirements

• 5+ years experience solving recognition problems in industry with launched products
• Fluency in C++ and Python
• Experience running ML models on mobile devices
• Experience with ML frameworks (PyTorch, Jax, TensorFlow)
• Skilled in software design, problem-solving, and debugging

Responsibilities

• Drive and solve semantic document parsing for customer use cases
• Participate in end-to-end quality improvements (metrics, prototyping, testing, integration, model creation, data collection, analysis, debugging)
• Stay up-to-date with latest research in document classification, text recognition, and information extraction
• Set direction and identify opportunities to push recognition boundaries
• Mentor team members on code quality and engineering practices