Data Scientist, Quality

Lilt is the leading AI solution for enterprise translations. Our stack made up of our Contextual AI Engine, Connector APIs, and Human Adaptive Feedback enable global organizations to adopt a true AI translation strategy, focusing on business outcomes instead of outputs. With Lilt, innovative, category-defining organizations like Intel, ASICS, WalkMe, and Canva are using AI technology to deliver multilingual, digital customer experiences at scale.

As a result, Lilt’s AI technology foundation is similar to ChatGPT and Google Translate, before our patented Contextual AI Engine, connector-first approach, and human-adapted feedback.

Our team is located globally in San Francisco, Indianapolis, London, and Berlin with hubs located in Washington D.C., New York City, and Boston. You can learn more about us and what it’s like to work here on our Careers page!

Authorization to work in the U.S. is a precondition of employment.

The Quality Analytics Team at Lilt

Lilt is a high-performance, large-scale language translation system powered by interactive machine translation. We have a large and diverse translation workforce that is distributed worldwide, whose interactions with the system form a fascinating large-scale dataset for researching translators' behavior.

Lilt’s Quality Analytics Team is part of the larger Production team which is a global workforce operating across four continents. We are passionate about localization and making information accessible to all. The team consists of project management, program management, quality, and talent management, working with some of the world’s largest organizations. Working on Lilt’s Quality Analytics Team you will have access to leading tools, collaborative teammates, and opportunities to make a real impact.

What you’ll do

You will be playing a key role in building the capabilities in language translation quality measurement and drive improvements in the quality deliverables. You will be building quality measurement tools to evaluate and monitor the quality of the translation of our human translators which would be used to make decisions on optimizing the document workflow. To develop these tools you will also be evaluating the use of various state-of-the-art tools including the latest developments in large language models. You will also be developing tools to assess the quality of the linguists that is most appropriate for a particular project. Additionally you will build predictive models to estimate the quality of the translation and enable optimal placement of the projects to maximize quality.

This position can be based out of any of our hubs and will be expected to work in the office in a hybrid capacity (where and when applicable).

Key Responsibilities

• Work closely with the Production, Product, and Engineering teams to develop methodologies to evaluate Quality of Translation output generated by Translators

• Develop complex machine learning models related to: 

  • Quality-estimation to estimate the quality and speed of translators

  • Driving improvement in Translator performance using our proprietary information

  • Predicting high quality performance based on leading indicators

  • Pricing in order to quickly and cost-effectively distribute work to Translators

  • Anomaly-detection to detect suspicious activity by Translators

• Use performance data to identify and categorize opportunities for improving quality of verified translations 

• Use quality measurement tools to compare quality of machine translation vs. human verified translation

• Develop the optimal design of human and automated quality measurements to maximize quality while managing review costs

Skills and Experience

• Masters or Ph.D. in Computer Science, Statistics, Applied Mathematics, Data Science, Operations Research, Artificial Intelligence, Human-Computer Interaction, or related disciplines

• 2+ years of professional experience working as a Data Scientist

• Experience developing machine learning models, especially for modeling users and predicting user behaviors, detecting anomalies, and adapting to users

• Exposure to the localization industry or language translation, with familiarity of the issues faced by translators, is highly desirable.

• Experience implementing machine learning systems in Python

• Experience implementing applications with Large Language Models in Python

• Experience working with large datasets about user behaviors and running statistical tests and analyses to gain insights from them

• Experience in quantitative user experience research methods, such as experiment