Job Search and Career Advice Platform

Ativa os alertas de emprego por e-mail!

Code Reviewer for LLM Data Training (SQL)

G2i Inc.

Fortaleza

Teletrabalho

BRL 269.000 - 377.000

Tempo integral

Ontem
Torna-te num dos primeiros candidatos

Cria um currículo personalizado em poucos minutos

Consegue uma entrevista e ganha mais. Sabe mais

Resumo da oferta

A leading AI training company in Brazil is seeking a Code Reviewer with SQL expertise to join their QA team. The role involves auditing AI-generated SQL responses and ensuring their technical correctness. Ideal candidates should have 5-7 years of SQL experience, strong analytical skills, and proficiency in English. This position offers flexible, remote training with competitive compensation.

Qualificações

  • 5–7+ years of experience in SQL development, QA, or code review.
  • Strong understanding of core SQL concepts and development best practices.
  • Experience with debugging, testing tools, and code execution environments.
  • Strong analytical and critical thinking skills.
  • Excellent written communication for documenting evaluations and feedback.
  • English proficiency at B2, C1, C2, or Native level.

Responsabilidades

  • Audit annotator evaluations of AI-generated SQL responses.
  • Verify if responses follow prompt requirements regarding query logic, joins, and correct SQL syntax.
  • Evaluate SQL code for correctness, performance, security, and readability.
  • Run and validate proof-of-work code submitted by annotators to confirm functionality.
  • Ensure responses align with instruction-following expectations including style, tone, and clarity.
  • Identify and document errors, omissions, or rating inconsistencies.
  • Provide clear, constructive QA feedback based on rubric and guideline criteria.
  • Collaborate with internal teams for clarifications on complex or ambiguous items.

Conhecimentos

SQL Development
Analytical Skills
Critical Thinking
Written Communication
Descrição da oferta de emprego

About the Company

G2i connects subject-matter experts, students, and professionals with flexible, remote AI training work such as annotation, evaluation, fact-checking, and content review. We partner with leading AI teams, and all contributions are paid weekly once approved, ensuring consistent and reliable compensation.

About the Role

We are looking for a Code Reviewer with expertise in SQL to join our AI training QA team. You will be responsible for auditing annotator evaluations of AI-generated responses written in SQL. Your role will be to ensure technical correctness, instruction adherence, proof-of-work validation, and consistent application of evaluation rubrics.

Responsibilities

  • Audit annotator evaluations of AI-generated SQL responses
  • Verify if responses follow prompt requirements regarding query logic, joins, and correct SQL syntax.
  • Evaluate SQL code for correctness, performance, security, and readability
  • Run and validate proof-of-work code submitted by annotators to confirm functionality
  • Ensure responses align with instruction-following expectations including style, tone, and clarity
  • Identify and document errors, omissions, or rating inconsistencies
  • Provide clear, constructive QA feedback based on rubric and guideline criteria
  • Collaborate with internal teams for clarifications on complex or ambiguous items

Required Qualifications

  • 5–7+ years of experience in SQL development, QA, or code review
  • Strong understanding of core SQL concepts and development best practices
  • Experience with debugging, testing tools, and code execution environments
  • Strong analytical and critical thinking skills
  • Excellent written communication for documenting evaluations and feedback
  • English proficiency at B2, C1, C2, or Native level

Preferred Qualifications

  • Experience with AI/LLM workflows, human-in-the-loop QA, or model evaluation
  • Familiarity with structured code evaluation processes and rubrics
Obtém a tua avaliação gratuita e confidencial do currículo.
ou arrasta um ficheiro em formato PDF, DOC, DOCX, ODT ou PAGES até 5 MB.