Senior Data Scientist

  • Jornada Completa
  • LinkedIn
  • Málaga

EPAM Systems

Senior Data Scientist

Acerca del empleo

Striving for excellence is in our DNA. Since 1993, we have been helping the world’s leading companies imagine, design, engineer, and deliver software and digital experiences that change the world. We are more than just specialists, we are experts.

EPAM is committed to providing our global team of 36,700+ EPAMers with inspiring careers from day one. EPAMers lead with passion and honesty and think creatively. Our people are the source of our success and we value collaboration, try to always understand our customers’ business, and strive for the highest standards of excellence. In today’s new market conditions, we continue to support operations for hundreds of clients around the world remotely, with the vast majority of our teams working from home. No matter where you are located, you’ll join a dedicated, diverse community that will help you discover your fullest potential.

Description

We are extending the team with Senior Data Scientists and Engineers. The role supposes participating in Conversational User Interfaces, we are using one of the Artificial Intelligence branches called Natural Language Understanding (or NLU).

To use NLU, we should follow a quite typical approach of:

  • Selection of real cases, in our case conversations, that reflect the reality of the use case we want to automate
  • Use a part of those cases to train and generate the NLU, creating what is called an NLU model. This NLU model is tailored to the concrete use case, and you want to make sure that it is not too narrow (so has too many similar data that only will respond to that specific phrases) and not too generic (has so much dispersion that responds to everything)
  • Another part of those cases will be used to test the NLU model, so we can ensure the NLU model can respond to our expectations, and the dataset has not to be part of the NLU model training and include examples to test the broader applicability of the model (thus containing examples that are not mapped in the model generation)

Required Skills

  • Data

Responsibilities

  • Analyze in deep the selected conversations to define uses cases or refine implementation by the designers
  • Map and measure the conversations flows (with several KPIs for the designers)
  • Identify flaws in the conversation design analyzing chatbots conversation with humans

Requirements

  • We expect NLP (Natural Language Processing) expertise in order to create the algorithms to:
    • Select/Identify the right conversations
    • Group and score similarities
    • Select/Unify training and test conversations
    • Identify key performance indicators such as sentiment, fulfilment performance and provide feedback to designers
  • We expect Big Data and Data Science expertise to:
    • Process all the actual available conversations across a whole year (each month we have hundreds of thousands of conversations)
    • Apply the algorithms in a), b) and c) in a scalable way, then can be automated and maintained
    • Collaborate in the implementation of automated supervised training based on conversations
    • Deep analysis on conversations crossed with other factors: stationarity, Contract-ending, new launches, Mktg campaigns, etc
    • Develop tools and dashboard for bot specific KPI tracking

We offer

  • Extended opportunity to grow professionally in a cross-cultural environment
  • Access to various on-line courses from leading provider
  • Access to engineering communities on a global scale
  • Unlimited access to LinkedIn learning solutions
  • Social benefits in line with local legislation
  • Health insurance and meal vouchers programs
  • Special discount program for EPAMers with providers across Malaga and in other cities around the world
  • Regular team collaboration events
  • Office in a good location with easy access
  • Referral bonuses
  • Relocation support (for people from other countries)

Additional

  • #DA_SP

Apply

Por favor, para apuntarte a este trabajo visita www.linkedin.com.