NLP Research researcher

In this role, you will join the Natural Language Processing (NLP) group at the Artificial Intelligence Research Institute and Sber AI Lab . Within the Institute, you will work together with researchers and engineers on the most challenging frontier scientific and technical problems, conduct experiments with state-of-the-art NLP models, and develop prototype systems that will be adopted in various products of the Sber ecosystem. You will need to engage with research topics, cover new domains quickly, build deep expertise with modern deep learning and natural language processing techniques. You will gain valuable experience in natural language processing, publish academic papers, and help push forward the quality of natural language understanding systems in one of the biggest financial and technological companies in Europe.

NLP RESEARCH ENGINEER RESPONSIBILITIES

• Develop and evaluate methods based on state-of-the-art machine learning and neural network methodologies for solving natural language processing tasks.
• Replicate methods and experiments proposed in recently published academic publications.
• Develop prototype AI-empowered systems that will be adopted in various products of the Sber ecosystem.
• Participate in the preparation of publications for the premier conferences in NLP and other fields of AI.

MINIMUM QUALIFICATIONS

• 2+ years of experience in a full-time software engineering position at a company or a research lab.
• 1+ year(s) of experience with an emphasis on deep learning.
• M.S. degree in Computer Science or a related field.
• Good English reading skills.
• Profound experience in developing and debugging in Python.
• Good knowledge of deep learning techniques. The ability to train and evaluate state-of-the-art deep learning models.
• Good knowledge of Python libraries: PyTorch, Numpy, Scipy, Pandas, Sklearn.
• Familiarity with natural language processing methods (at least on the basis of massive open online courses, e.g. Coursera).
• Fluency with Git, Bash, Linux, and other basic software development tools.     

PREFERRED QUALIFICATIONS

• M.S. degree in Computer Science or a related field.
• 2+ years of experience with an emphasis on deep learning.
• Good English reading and writing skills.
• Profound experience in building systems based on machine learning and deep learning.
• Experience in building natural language processing systems: text classification, machine translation (MT), sentiment analysis (SA), information extraction (IE), relation extraction, semantic parsing (SRL), sequence tagging, especially named entity recognition (NER), dialogue systems.
• Experience in using deep learning frameworks: PyTorch, Tensorflow, and Keras.
• Experience in using NLP packages: HuggingFace Transformers, AllenNLP, SpaCy,
• NLTK, Flair, Gensim, word2vec, doc2vec, fastText, DeepPavlov, GluonNLP, etc.
• Experience in using machine learning packages: XGBoost, CatBoost, LightGBM, FastAI, etc.
• Experience in working with Docker/Singularity/Podman.
• Experience with the following models and algorithms: CRF, LSTM, GRU, Transformer, Seq2Seq, attention, CNN, SGNS, ELMo, BERT, DeBERTa, KnowBERT, LUKE, etc.
• Experience with Bayesian deep learning models and related techniques.
• Experience with imitation learning and reinforcement learning methods.

We offer a competitive salary depending on the experience and background, a working contract according to the Russian labor regulations, and the possibility to work remotely a couple of days a week if needed.