— Python programming language.
— Experience with algorithms and machine learning libraries in NLP (with some of the presented ones): nltk, Stanford NLP toolset (parser, NER, coreference resolution, word segmenter e.t.c.) spacy, genism, and bigartm.
— Experience with Russian and Kazakh languages:
— morphology: pymystem3, pymorphy2
— grammar parsers: Tomita parser, yargy
— syntax: udpipe and \ or syntax net and others.
— Experience in building deep neural networks using Tensorflow, Keras, PyTorch frameworks.
— Version control: git.
— OS: Linux.
— Experience in building models using optimal embeddings, memory, and attention.
— Experience with ipavlov and alennlp; understanding of the peculiarities of the Russian language.
— Experience in testing and debugging methods/concepts for processing textual data, working with markup Universal Dependencies and similar.
— Development of dialogue systems.
— Knowledge of English
— Opportunity to apply your knowledge to work on interesting and international projects;
— Work in a developing ambitious company in a team of professionals;
— Full-time work. 10.00 to 19.00
— Text preprocessing.
— Classification, clustering of texts.
— Extract named entities and keywords.
— Morphological, syntactic analysis.
— Revealing intents.
— Thematic modeling.
— Spell checker.
— Information search and duplicate detection.
— Distribution semantics: word2vec / paragraph2vec, fasttext, etc.
— Analysis of unstructured text communication
— Working with external sources (arxiv, github)
— Measure / monitor the work of markers.
Our partner develops innovative products for pharmaceutical retail to improve the efficiency of interaction between consumers and manufacturers. It uses technologies in CV & Speech-to-Text with subsequent natural language processing and data visualization to improve the quality of service and customer satisfaction. They have successfully passed the program from Nvidia Inception and are ready for global challenges. Now they have 2 teams.