Our team helps develop Sber's ATM network and ensure its smooth operation. We are involved in various aspects of this process, from an AI agent in next-generation ATMs to automating the processing of device maintenance requests. The team works with both modern solutions based on LLM, RAG, and AI agents, as well as classic ML tasks.
Responsibilities
- Develop and enhance AI agents and LLM systems: build pipelines, experiment with models and approaches, evaluate quality, and improve solutions
- Solve classic ML and NLP tasks: classification, working with texts, embeddings, and other ML models
- Conduct experiments with models and their parameters, analyze results, and select the best approaches
- Develop data preparation and processing pipelines for ad-hoc analytics and data mart construction
- Organize and conduct quality assessment of ML/LLM solutions, work with data labeling
- Participate in the development of solutions at all stages: from task definition and customer communication to implementation and production monitoring
Requirements
- Understanding of AI agent construction principles
- Proficient in Python and SQL, ability to work with Git
- Understanding of the main stages of ML solution development
- Knowledge of classic ML fundamentals: classification, regression, main algorithms, and quality metrics
- Understanding of main NLP approaches and working with text data
Will be a plus:
- Knowledge of RAG systems
- Knowledge of Hadoop, Spark
- Basic understanding of MLOps (Docker, Jenkins) and Linux
Conditions
- Comfortable modern office near Kutuzovskaya metro station (office or hybrid work mode)
- Corporate gym and recreation areas
- Sber's unique training system for professional and career development, seminars, trainings, conferences
- Adaptation program and supervisor support at the start
- Voluntary medical insurance, accident insurance, social guarantees
- Flexible mortgage loan discount equal to 1/3 of the Central Bank's key rate
- Prime subscription with the possibility of shared use by three close individuals
- Reward for recommending friends to the Sber team
- Corporate pension program
- Corporate events.
