AI Safety Experts — English & Bengali
Bilingual Competency CORE Not done RLHF Safety Proficiency Assessment Not done Work Authorization Not done All application steps are reused whenever another role requires the same step, so you never have to upload your resume or take the same interview twice Location: Remote Fluent Language Skills Required: English & Bengali. Applicant pay shown: $20 - $22 / hour.
- Platform
- Mercor
- Fit category
- Language and bilingual
- Remote/location
- Remote
- Pay
- Firm hourly pay: $20 - $22 / hour
- Eligibility
- Remote, work authorization mentioned
- Difficulty
- Selective
- Last reviewed
- Jul 1, 2026
Current application
Check the current Mercor listing
Apply on MercorReview the official Mercor listing before applying. Requirements, screening, pay, hours, and project availability can change.
What this role involves
Bilingual Competency CORE Not done RLHF Safety Proficiency Assessment Not done Work Authorization Not done All application steps are reused whenever another role requires the same step, so you never have to upload your resume or take the same interview twice Location: Remote Fluent Language Skills Required: English & Bengali. Native fluency in English and Bengali is required for this position. Why This Role Exists At Mercor, we believe the safest AI is the one that’s already been attacked — by us. We are assembling a red team for this project - human data experts who probe AI models with adversarial inputs, surface vulnerabilities, and generate the red team data that makes AI safer for our customers. This project involves reviewing AI outputs that touch on sensitive topics such as bias, misinformation, or harmful behaviors.
Requirements
- You bring prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing)
- You’re curious and adversarial: you instinctively push systems to breaking points
- You’re structured: you use frameworks or benchmarks, not just random hacks
- You’re communicative: you explain risks clearly to technical and non-technical stakeholders
- You’re adaptable: thrive on moving across projects and customers
- Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction
Best for
- Domain specialists
- Careful reviewers
- Applicants comfortable with Mercor assessments
Skip this if...
- You need guaranteed acceptance, guaranteed work, guaranteed pay, or guaranteed hours
- You are not comfortable with Mercor's application, assessment, or independent-contractor terms.
- You require H-1B or STEM OPT support and the live listing still says those candidates are not supported.
Application tips
- Prepare concise proof of relevant professional experience before starting the application flow.
- Verify the live Mercor listing before applying because availability, pay, and eligibility can change.
- Read any location, work-authorization, and assessment requirements before applying.