As a data scientist you will push the boundaries of deep learning & NLP by conducting research in cutting-edge AI and apply that research to production scale data. Our data includes a large corpora of global patents in multiple languages and a wealth of metadata. The focus of your work will be to analyze, visualize, and interpret large text corpora as well as more traditional structured data. This includes recommending and implementing ML-based product features. We also expect to publish primary research and contribute to FOSS that we use.

This is an exciting opportunity to part of a startup that is applying deep learning to a real-world problem at global scale. We’re always looking for leaders and there is room to grow for the right person to take increasing responsibility working as part of a small and dynamic team.

Location:

Tokyo, Melbourne, San Francisco (willing to negotiate remote work as well)

 

Responsibilities:

• 
  Architect and implement software libraries for batch processing, API-based predictions, and static analyses
  


• 
  Rapidly iterate on the design, implementation and evaluation of machine learning algorithms for document corpora
  


• 
  Report and present software developments including status and results clearly and efficiently, verbally and in writing
  


• 
  Participate in strategy discussions about technology roadmap, solution architecture, and product design
  


• 
  Strict adherence to clean code paradigms
  

• 
  BSc/BEng degree in computer science, mathematics, machine learning, computational linguistics or equivalent (MSc/MEng preferable)
  


• 
  Experience with implementing statistical methods and data visualization
  


• 
  Good knowledge of computer science principles underpinning the implementation of machine learning algorithms
  


• 
  Experience with deep learning approaches to NLP, particularly RNNs
  


• 
  Experience implementing deep learning models in TensorFlow or PyTorch
  

• 
  Contributions to open source projects
  


• 
  Passion for new developments in AI
  


• 
  Experience with GCP or AWS
  


• 
  A track record of machine learning code that is:
  
  
  
  
  • 
    Well documented
    
  
  
  • 
    Well commented
    
  
  
  • 
    Version controlled
    
  
  
  • 
    Unit tested