Description
Our data labeling team helps train artificial intelligence. Training neural networks requires massive datasets. These datasets need to be classified, annotated for key elements, evaluated, and more.
We are growing and looking for a specialist with a good command of the Russian language to prepare data for GigaChat and other neural networks.
Responsibilities
What we offer:
- work in the office at: Oktyabrsky pr.23/25, possibility to transition to a remote schedule based on performance
- full-time employment, schedule 5/2 from 09:00-18:00, Friday until 16:45
- salary plus a bonus system based on performance
- formal employment according to the Labor Code of the Russian Federation
- voluntary health insurance and financial support
- corporate training
- career growth
- vibrant and active corporate life (holidays, competitions, sports events)
- discounts on Sberbank and ecosystem products
- flexible discount on mortgage loans, equal to 1/3 of the Central Bank key rate
- referral bonus for recommending friends to the Sber team.
Requirements
What you will do:
- evaluate user queries and model responses from a safety perspective
- check model responses for stylistic and grammatical errors
- classify text files according to established rules
- verify the accuracy of all facts in model responses
- select the best model response among many others.
Before each series of tasks, there are instructions and training examples that will help you understand the task in more detail.
Conditions
We expect you to:
- be a specialist with a higher education degree
- be attentive, diligent, not afraid of routine, and ready to analyze a large volume of data
- be knowledgeable — able to understand queries on absolutely any topic: from poetry to politics
- be ready to use AI in daily work (company-funded training — we have our own courses).
While completing the test assignments, you will have the opportunity to try on the role of a data labeling specialist and see how interesting it will be for you.
