Get Started
Hunting for you, 24x7Get the iPhone app

Your hunt runs in your pocket, and tells you the moment a strong match opens.

Download on theApp Store
CleraDate not stated

Safety Testing Associate

Worldwide remote — Open worldwideEntry levelRemoteFull-time$30 – $60 an hour
Who we'd reach at CleraFound and reached by Power Apply
Best leadHiring manager

Leads the team this role joins

Recruiter

In-house, owns the opening

Head of the team

Director or head of the group

Power Apply

From your own Gmail and LinkedIn.

About the role

This is a fully remote, full-time safety testing role open to applicants worldwide, with 0-5 years of experience. It suits someone who can try to break AI systems, document the results and judge outputs against written safety policies. The role sits in an expert network that supports AI model evaluation projects. Projects are matched to each person's background.

What you’ll do

  • Write prompts and multi-turn conversations to test whether models can be pushed past safety guidelines.
  • Test known and new jailbreak methods and record which succeed.
  • Check whether embedded instructions in documents or tool outputs make a model follow unintended directions.
  • Assess whether models with code, browser or email tools can be steered into harmful actions.
  • Compare model responses across groups, names and languages to spot bias.
  • Audit answers against written policies and flag violations.
  • Label content as safe or unsafe under set guidelines.

What they’re looking for

  • 0-5 years of professional experience. Prior AI data work is not required.
  • Experience testing software or models by trying to break them and documenting each step.
  • Experience applying written policies or guidelines to content or behaviour.
  • A background in security testing, CTFs, trust and safety, content moderation, policy analysis, fairness research, adversarial ML, QA, or cyber, bio or chemical expertise.
  • Clear written English and careful documentation.
  • Ability to handle sensitive content under structured guidelines.

Nice to have

  • Knowledge of jailbreaks, prompt injection or red teaming of language models.
  • Experience with Burp Suite, Garak or PyRIT.
  • Familiarity with the OWASP LLM Top 10.

What’s on offer

  • Pay of USD 30 to 60 per hour, depending on project, field and experience.
  • Payment is based on hours worked.
  • Fully remote work.

Skills

Red TeamingPrompt InjectionJailbreakingLarge Language ModelsLLM EvaluationPolicy Compliance ReviewContent ModerationBias TestingBurp SuiteGarakPyRITOWASP LLM Top 10Technical Documentation

Questions about this role

Is this role remote?

Yes. It is fully remote and open to applicants worldwide.

How much does it pay?

USD 30 to 60 per hour, paid for hours worked.

Do I need AI experience?

No. Prior AI data work is not required, though familiarity with prompt injection or red teaming helps.

How many years of experience are needed?

Zero to five years of professional experience.