English and Kannada AI Safety Red Team Reviewer

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. OpenTrain AI hires and contracts contributors for projects that help improve modern AI systems, while giving you a profile to showcase your experience and build a lasting portfolio.

• Create a free OpenTrain account and apply in minutes.
• Build a credible record of AI training and evaluation experience.
• Find flexible projects that match your language and analytical skills.

About AI Safety Training

AI training is the human work behind better artificial intelligence. Reviewers examine model behavior, identify risks, and provide structured feedback that helps AI systems become more accurate, useful, appropriate, and safe.

Red teaming is a specialized form of AI evaluation. Contributors deliberately probe models with challenging or adversarial interactions so technical teams can understand weaknesses before they affect users. This role focuses on text-based conversational AI in English and Kannada.

• Remote work in a fast-growing area of technology.
• Human feedback can shape how conversational AI behaves.
• Flexible project work can fit around other commitments.

The Role

OpenTrain AI is seeking an English and Kannada Conversational AI Safety Red Team Reviewer. You will test conversational AI models and agents, assess their responses, and produce structured human feedback about safety vulnerabilities and model behavior.

This is an entry-level contractor role requiring at least 20 hours per week. The pay range is $16 to $22 per hour, and the engagement is part time.

• Role type: Part-time contractor
• Experience level: Entry level
• Time requirement: 20+ hours per week
• Pay: $16-$22 per hour
• Primary data type: Text
• Languages: English and Kannada

What You’ll Do

You will challenge conversational AI systems with adversarial prompts and multi-turn interactions. Your work will help identify failures that may be difficult to detect through ordinary testing, then turn those findings into clear, reproducible records for technical and non-technical audiences.

• Probe models and agents for jailbreaks and prompt injections.
• Test misuse cases, bias exploitation, and multi-turn manipulation.
• Review outputs for accuracy, completeness, appropriateness, and safety.
• Annotate failures and classify identified vulnerabilities.
• Apply taxonomies, benchmarks, playbooks, and testing guidelines consistently.
• Flag systemic risks and subtle errors or inconsistencies.
• Document reproducible attack cases, reports, and datasets.
• Explain reasoning clearly so teams can understand model weaknesses.

Requirements

You must be natively fluent in both English and Kannada and have strong written communication skills. Careful language judgment, rigorous analysis, and sound decisions about conversational AI behavior are central to the role.

You should be comfortable working independently, following structured quality standards, and adapting across different task types. Experience in adversarial machine learning, cybersecurity, abuse analysis, or conversational AI testing is useful.

• Native fluency in English and Kannada.
• Strong written communication and analytical judgment.
• Ability to identify jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation.
• Ability to assess whether AI responses are accurate, complete, appropriate, and safe.
• Ability to classify vulnerabilities and apply structured standards consistently.
• Useful experience with adversarial machine learning, cybersecurity, abuse analysis, or conversational AI testing.
• Ability to explain findings clearly to technical and non-technical audiences.

Who Can Apply

This opportunity is available to contractors in Austria, Belgium, Bulgaria, Canada, Cyprus, Czechia, Germany, Denmark, Estonia, Spain, Finland, France, the United Kingdom, Greece, Croatia, Hungary, Ireland, Italy, Lithuania, Luxembourg, Latvia, Malta, the Netherlands, Poland, Portugal, Romania, Sweden, Slovenia, Slovakia, and the United States.

• Applications are limited to the countries listed for this role.
• English and Kannada fluency are both required.

How to Apply Through OpenTrain

Create a free OpenTrain account, build your profile around your language and AI safety evaluation skills, and apply to the role in minutes. Your OpenTrain profile can help you demonstrate relevant work and grow your AI training career over time.

• Set up your OpenTrain profile.
• Highlight English and Kannada fluency and relevant safety-testing experience.
• Apply for this contractor opportunity through OpenTrain.