Quality Assurance Specialist

Chagible; United States; A distancia; USD 34.000 - 83.000; A tiempo parcial

Descripción de la vacante

Chagible is seeking a remote, independent contractor to rigorously test AI-powered applications. You will evaluate prompts, model responses, and multi-turn conversations, identifying inaccuracies, bias, and safety concerns.

Collaboration with engineers and prompt designers ensures fixes before and after release. You will document issues with reproducible steps, review training data quality, and support API integrations and performance monitoring.

Formación

• Hands-on experience testing generative AI, language models, or conversational apps.
• Familiarity with prompt and output evaluation, including multi-turn interactions.
• Ability to spot inaccurate, inconsistent, biased, or unsafe AI responses.
• Experience working with AI service APIs or model integrations and reviewing logs or performance indicators.
• Basic understanding of natural language processing and language model behavior.
• Careful, systematic approach to testing, edge cases, and issue documentation.

Responsabilidades

• Run structured tests across prompts, response flows, and unusual or high-risk scenarios.
• Assess responses for accuracy, consistency, relevance, bias, hallucinations, and safety concerns.
• Check application behavior before and after deployment, including errors, logs, response stability, latency, and token use.
• Record findings with the relevant prompts, outputs, and clear steps for reproducing each issue.
• Review the quality and consistency of AI-generated training data and annotations.
• Collaborate with developers and prompt engineers to validate fixes and improvements.

Conocimientos

AI testing

Prompt evaluation

Bias and safety spotting

APIs and logging review

NLP basics

Documentation

Descripción del empleo

Role overview

Help ensure AI-powered applications behave reliably and safely in everyday use. You’ll test prompts, model responses, and multi-turn conversations, then work with engineering and prompt design colleagues to address issues before and after release.

Responsibilities

• Run structured tests across prompts, response flows, and unusual or high-risk scenarios.


• Assess responses for accuracy, consistency, relevance, bias, hallucinations, and safety concerns.


• Check application behavior before and after deployment, including errors, logs, response stability, latency, and token use.


• Record findings with the relevant prompts, outputs, and clear steps for reproducing each issue.


• Review the quality and consistency of AI-generated training data and annotations.


• Collaborate with developers and prompt engineers to validate fixes and improvements.

Requirements

• Hands-on experience testing generative AI, language models, or conversational applications.


• Familiarity with prompt and output evaluation, including multi-turn interactions.


• Ability to spot inaccurate, inconsistent, biased, or unsafe AI responses.


• Experience working with AI service APIs or model integrations and reviewing logs or performance indicators.


• Basic understanding of natural language processing and language model behavior.


• Careful, systematic approach to testing, edge cases, and issue documentation.

Benefits and work setup

Remote independent-contractor role for candidates based in Singapore or the Philippines.