Data Scientist / ML Researcher

by Mait Müntel | Jobs | 21 November 2019

Do you feel you could have a greater impact on the world as a young PhD? Do you feel that you’ve worked hard for decades to build up your knowledge and skills and that you’d like to put them to use in the real world? If you’d like to work on tough problems that improve the lives of millions of people, do the research, and be competitively compensated, please read on.

Our goal is to revolutionise learning

Education is the single biggest investment any person can make, but we’re still learning as we did 100 years ago. The world is changing fast, and the increasing use of AI is expected to impact the job market. Many jobs as we currently know them risk becoming redundant. We will not be able to cope with these changes if we keep learning at the same pace, using the same methods we used 100 years ago.

That’s why we at Lingvist are building technology to create a step change in human learning.

It is not an easy challenge, but problems worth solving seldom are. And we believe that we can do it with the help of ML/AI and brilliant scientists.

Who we are

Lingvist began as a prototype that was built at CERN as a hobby in order to accelerate language learning using machine-learning algorithms. Soon after, the technical co-founder of Skype and its core technology leader joined because they saw the potential for building another high-impact company. Since that time, Lingvist has attracted US$15m in investments from the Japanese e-commerce giant Rakuten, Skype co-founder Jaan Tallinn, and different Nordic VC firms. Recently, the University of Tartu measured the impact of Lingvist and determined that it can help people learn significantly faster when compared to competing commercial solutions.

We are committed to the research that enables us to further develop our core technology. We’ve now reached 3 million users and will grow much more in the coming years. Our user base helps us to run experiments and train our ML models. We believe this will enable us to push the boundaries of human learning far beyond what is currently possible.

We started off with languages because they comprise the biggest unique segment in education. For example, there are 1.5 billion people currently learning English in order to gain access to better education and job opportunities. This presents a great opportunity to scale out our core technology to other fields of education in the future.

Your role

We are seeking a Data Scientist / ML Researcher. This person would join our cross-functional research team, a group with skills ranging from NLP and cognitive science to engineering and design, which is ready to test the learning models and algorithms you develop on real users, with fast iterations. Your role will be to help formalise the learning models so they are measurable, interpret the user data resulting from these experiments, and play a key role in improving our learning algorithms.

Your responsibilities

  • Put your model-building skills to work in solving the problem of human learning by building computational models for human memory and language acquisition.
  • Quantify the learning process so that it can be measured and analysed in detail to make systematic progress towards improving learning efficiency.
  • Design, create, and train computational models to predict the learning value of user actions. Build algorithms to serve the most optimal sequence of those actions to the user in order to maximise knowledge acquisition efficiency.
  • Work closely with NLP and learning experts, engineers, and designers to create experiments that enable better insights into human learning and improve our technology.
  • Analyse the detailed data of millions of users (~10TB today, ~PB in the future), present it to the other team members, and help to interpret it based on the models we’ve developed.
  • Help direct the future – what new data should we collect, how should we look at data, and where else could we be creative?


  • Great model-building skills – you love to build models, measure them against the real world, and improve iteratively. You have experience using abductive reasoning methods and building computational models, and you have earned your PhD in a field where those skills are practised (physics, cognitive psychology, biophysics, or similar).
  • Great coding skills – you are fluent in your favourite programming language. You know Python well enough (or can learn it quickly) in order to use it as a tool and have extensive experience using several ML/AI and statistical tools.
  • Good communication skills – you can work efficiently in a team, because you understand that teams outperform individuals. You’re good at written, remote, and face-to-face communication in English. Great communication skills are especially important for those who see themselves as team leaders at some point in their career at Lingvist, since a young and growing company will provide such opportunities.
  • A passion for languages and/or learning – you personally value learning. As we’re currently focusing on languages, it would help tremendously if you wanted to learn a language on your own with the technology you’re building. This will give you an inside advantage in using your model-building skills to the fullest.

What we can offer you

  • Exciting work on a challenging problem that impacts the world positively at scale
  • A commitment to supporting your self-development and career ambitions
  • The opportunity to work together with the world-class engineering team that built Skype
  • Open and honest communication and feedback
  • A working environment filled with talented people with high integrity
  • A competitive salary to let you focus fully and enjoy what you do

Job specifics

  • Full-time (permanent) contract
  • Location: London, Tallinn, Munich, Barcelona, or a remote arrangement in Europe with regular trips to Tallinn

If Lingvist sounds like a company you’d be interested in joining, then please fill out the form here.