Description
Our team develops smart solutions and reporting for the internal needs of teams. In our solutions, we use both classical models and AI agents based on the LLM GigaChat. To complete tasks, we collect data, prepare prototypes, and validate the developed tool. Our goal is to provide our users with quality tools for solving issues of data processing and analysis, etc., using advanced AI technologies, including Generative AI.
Responsibilities
- hypothesis testing
- dataset preparation
- development of AI agents on LangGraph + GigaChat, prompt engineering, fine-tuning LLM
- development of RAG pipelines
- model development (classification, recommendations, predictions)
- evaluation of model quality metrics
- development of api-services.
Requirements
- work experience as an ML engineer from 1.5 years
- confident programming in Python
- experience interacting with LLM
- ability to evaluate the quality of LLM-based solutions
- use and development of REST API (aiohttp, flask)
- work with databases (PostgreSQL) using ORM.
Will be a plus:
- experience participating in machine learning/Data Science competitions or hackathons
- github profile
- experience solving algorithmic problems (Leetcode, Codewars, Hackerrank, etc.)
- skills in working with RAG pipelines
- experience in creating AI agents
- work with GigaChat
- experience working with the GreenPlum DBMS.
Conditions
- hybrid work format
- office at the address: Moscow, Kutuzovsky Prospekt, 32k1
- more than 400 educational programs from SberUniversity for professional and career development
- flexible discount on mortgage loans, equal to 1/3 of the Central Bank's key rate
- extended voluntary health insurance (VHI), preferential insurance for family and corporate pension program
- free subscription to SberPrime+, discounts on products of partner companies
- reward for recommending friends to the Sber team
