Bixby is an intelligent virtual assistant which is available as a built-in application on Samsung flagship devices. This application uses a natural language user interface to perform tasks on these devices using voice, including but not limited to:
– Making phone calls.
– Sending text messages.
– Setting up meetings.
– Opening apps.
– Setting alarms and timers.
– Getting directions.
– Answering general questions.
– Providing rich services such as find restaurants.
– Book hotel by Hotels.com API.
– Order Starbucks coffee.
– Call Uber and extend additional service with outside content providers.
Samsung R&D Institute Jordan office is currently seeking Bixby capsule developers and Natural Language (NL) developers. The capsule is a unit of developer’s contributions which consisting of the intended service’s logic, UI, various workflows and dialogs, and Natural Language training data. As a capsule developer, you will implement and maintain different types of capsules. Some are associated with a downloaded Android application (e.g. preloaded Call app, Camera app), while some are fully residing in the cloud (most Content Provider-based services). There are some pre-built capsules that provide shared services to other capsules, called library capsules. As an NL developer, you will implement and maintain each language specific tokenizer. The NL developer will also be responsible for providing NL training guides to the capsule developers by reviewing the models, annotations and other aspects of the capsule to best handle NL processing. In Bixby’s decentralized, scalable approach, the quality of the service is dictated by the quality of smart training data of each capsule.
• Capsule development (soon to be open to 3rd party developers)
o Implement object-oriented concept and action modeling.
o Map JS actions with REST API calls
o Creating a conversation with dialogs between the user and the capsule.
o Creating visual representation of the UI by utilizing system-defined widgets or by creating HTML5/CSS layouts.
o Natural language training by breaking natural utterance to modeled units.
o Translation and migration of data and annotations done in English to each language.
• Library capsule development (internal capsules)
o Import and manage geo database written in local language for geo capsule localization.
o Implement geo actions using 3rd party location & POI CP APIs.
o Setup and deploy local service utilizing AWS or similar cloud services.
• Tokenizer & Internationalization
o Develop and maintain language-specific tokenizer.
o Provide guideline for capsule developers on writing distinct training data minimizing quantity (smart data over big data) in a scalable approach.
|Job Location:||Amman, Jordan|
|Job Role:||Research and Development|
|Company Industry:||Information Technology|
|Career Level:||Mid Career|