Назад
Company hidden
10 дней назад

Data Annotator (SQL)

Тип работы
fulltime
Грейд
middle
Английский
b2
Вакансия из списка Hirify.GlobalВакансия из Hirify Global, списка международных tech-компаний
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/

TL;DR

Data Annotator (SQL): Analyzing and refining user-chatbot interactions for an internal corporate analytics product with an accent on dialogue quality and data accuracy. Focus on identifying problematic scenarios, building query datasets, and optimizing chatbot performance using SQL and internal admin tools.

Company

hirify.global is a technology service provider facilitating expert talent placement.

What you will do

  • Analyze user queries and system dialogues within the admin panel and Langfuse to monitor request processing pipelines.
  • Identify and document problematic interaction scenarios to improve system reliability.
  • Enrich the dataset of user queries for few-shot learning and quality evaluation.
  • Write and refine SQL queries to verify data correctness and improve chatbot responses.
  • Contribute to the improvement of admin panel functionality and overall product workflow.

Requirements

  • 3+ years of experience in data, dialogue, or content annotation.
  • Ability to analyze complex user-system interaction scenarios.
  • Strong attention to detail, structured thinking, and ability to follow guidelines.
  • Basic SQL proficiency including SELECT, JOIN, and WHERE queries.
  • Understanding of table structures and ability to adjust existing SQL queries.
  • Willingness to work with Langfuse and proprietary admin tools.

Nice to have

  • Experience working with ML products or chatbots.
  • Understanding of LLM response quality evaluation principles.
  • Experience in creating few-shot sets and test query libraries.

Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →