Speech Recognition Researcher & Developer

Ramat Gan · Full-time

About The Position

Shield is a fast-growing Start-Up in the domain of Electronic Communication Compliance, solving complex problems within enterprises, using a blend of regulatory expertise, technology, and artificial intelligence. Our platform analyzes huge amounts of electronic communications to fight financial crimes.

It is an exciting time to join with much growth ahead. We have successfully completed a Series A investment round and will be growing rapidly in the near future. Shield is a special and limitless place to work where individuals are encouraged to bring their passion and align to our shared purpose and culture of excellence and innovation. 

We are looking for a highly motivated Data Scientist to start our journey into the speech recognition world. If you love solving a big problem of speech recognition in the financial world, have prior experience in NLP & Voice technology we are the place you want to grow with.

in this role, you will join our Data Science Team and be responsible for developing a speech recognition engine that is custom built to our needs. You will use unique data which we already have as part of our data platform in order to enhance the accuracy of the voice transcription. This development is critical to Shield's ability to find the needle in a haystack in the world of structured & unstructured data of electronic communication. 


  • MSc or PhD in Electrical Engineering or Computer Science or a related technical field
  • Minimum 5 years of experience in development of data driven algorithms for speech and language processing
  • Technical experience in one or more of the domains below

Speech recognition technologies

Data driven natural language processing

Machine learning platforms such as TensorFlow or Cognitive Toolkit

Deep Learning

  • Deep familiarity in C++ & Python
  • Creative mind and high analytical skills
  • Ability to lead, take ownership of your territory and guide others

Apply for this position

Please, visit our site using another browser