We are a team focused on deep content understanding within VK's experimental technologies division. We are currently actively developing areas around the integration of LLMs into product solutions and multimodality. We are building systems that can understand video, images, text, and audio, extract information from them, and solve applied product tasks within the company's ecosystem. The team is growing rapidly, many areas are being built from scratch, so you can propose your own ideas and influence the development of the company's ML services.

Tasks

• Development and improvement of LLM and multimodal ML pipelines
• Working with video, text, images, and audio
• Integration of models into product mechanics

Requirements

• Good knowledge of Python
• Understanding of machine learning fundamentals
• Experience with deep learning frameworks (pytorch)
• Experience in Deep Learning
• Knowledge of SQL
• Understanding of NLP/LLM architectures
• Good knowledge of basic algorithms and data structures

Will be a plus

• Relevant projects or internships
• Experience in developing and implementing ML models
• Relevant education

Conditions

• Fixed-term employment contract for 6 months