Lead Machine Learning Engineer

DocuSign DocuSign · Enterprise · Dublin, Ireland · Engineering

Lead Machine Learning Engineer at DocuSign, focusing on building and deploying production-level AI/ML solutions for document understanding, leveraging computer vision and NLP. The role involves the full ML lifecycle, from research and development to deployment and monitoring of models within the Docusign Agreement Platform.

What you'd actually do

  1. Work together with your team to perform model development, research, testing, and evaluation of existing and emerging deep learning methods and technologies that can be effectively applied to the contract domain
  2. Apply the latest architectures and technologies to build Docusign IP and solve complex NLP challenges including, but not limited to, generating representations, text understanding, semantic retrieval, contextual extractions, summarization
  3. Apply computer vision for document understanding tasks such as layout and object detection, image classification and tagging
  4. Understand, assist, and improve the existing model training, evaluation, and online inferencing processes, define online metrics, and design user feedback for our AI / ML features
  5. Work closely with the engineering partners to deploy models into production, build scalable AI systems, monitor and improve performance metrics

Skills

Required

  • Experience in designing, developing, deploying and monitoring machine learning and deep learning solutions
  • Programming in PyTorch, TensorFlow, or equivalent deep learning framework
  • Fluent in Python
  • Bachelor’s degree in computer science, physics, statistics, econometrics, operations research, applied mathematics or an equal computational field
  • Experience in latest NLP techniques including LLMs and language representations
  • Experience in text extraction techniques, especially using OCR and direct extraction from docx, images, and pdfs

Nice to have

  • Experience developing and deploying large scale document understanding models in production
  • Hands-on experience building multi-modal solutions between CV or NLP
  • Extensive experience in data collecting, cleaning, sampling, and processing large, diverse structured or unstructured datasets

What the JD emphasized

  • production-level machine learning models
  • deploy models into production
  • build scalable AI systems
  • online inferencing processes
  • document understanding models in production
  • multi-modal solutions

Other signals

  • building production-level machine learning models
  • deploying models into production
  • build scalable AI systems