It's a new product of a company is the 3rd largest social discovery company in the world that unites people communication through online platforms focusing on AI, game mechanics, and video streaming.

Responsibilities:

• Solving NLP problems such as summarization, classification, clustering, NER, using modern techniques including LLM;
• Participate in the full development cycle, from problem definition to implementation;
• Building a data processing and pre-training LLM pipelines for text generation, chatbot development, RAG systems;
• Developing new approaches and data partitioning processes to assess the quality of LLM performance;
• Extracting data from various sources (reading from files, APIs, databases);
• Participate in generating requirements and necessary data for model improvements;
• Implementing the model in production, supporting the model lifecycle, monitoring and updating.

Requirements:

• Experience in the role of Data Scientist/LLM engineer for more than 5 years
• Experience in Data Science related to natural language processing (NLP);
• Proficiency in Python, including NumPy, Pandas, Scikit-learn and text processing libraries;
• Practical experience in developing and implementing NLP models for classification, text clustering, NER (Named Entity Recognition), etc;
• Experience with deep learning frameworks (TensorFlow/Keras or PyTorch), including model building, training and evaluation;
• Knowledge of GPT, BERT and other Transformer model architectures;
• Knowledge of various metrics for evaluating the quality of NLP models (precision, recall, F1-score, AUC-ROC, etc.), ability to select appropriate metrics for a particular task;
• Experience in bringing models to production;
• Experience in using tools for performance and quality monitoring;
• Experience with model training: Qwen, llamaindex and Mistral.