AI Safety Expert, English – Bengali
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
AI safety red teamer probing frontier AI models for Mercor. Testing adversarial vulnerabilities and producing human data, datasets, and actionable reports.
Responsibilities:
- Red-team conversational AI models and agents
- Test jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation
- Annotate AI failures, classify vulnerabilities, and flag systemic risks
- Apply taxonomies, benchmarks, and playbooks to maintain consistent testing
- Produce reproducible reports, datasets, and attack cases for customers
- Identify vulnerabilities that automated tests miss
- Expand evaluation coverage and reduce production surprises
- Help Mercor customers strengthen the safety, robustness, and trustworthiness of their AI systems
- Work on projects training and enhancing frontier AI systems
Requirements:
- Fluent/native fluency in English and Bengali required
- Prior red teaming experience in AI adversarial work, cybersecurity, or socio-technical probing
- Ability to probe AI systems adversarially, including jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation
- Ability to generate human data by annotating failures, classifying vulnerabilities, and flagging systemic risks
- Ability to follow taxonomies, benchmarks, and playbooks
- Ability to produce reproducible reports, datasets, and attack cases
- Ability to explain risks clearly to technical and non-technical stakeholders
- Adaptability across projects and customers
- H1-B and STEM OPT candidates are not supported
- Independent contractor engagement
Benefits:
- Fully remote role that can be completed on your own schedule
- Payments are weekly on Stripe or Wise based on services rendered
- Higher-sensitivity projects are optional
- Clear guidelines and wellness resources for higher-sensitivity projects
- Competitive pay
- Referral bonus of up to $90 for each successful referral
- No limit on referrals (restrictions may apply)
Similar Jobs
Data Privacy and Responsible AI Manager
Data privacy and responsible AI manager for Accuray’s cancer radiotherapy systems. Assessing privacy, vendor, AI, and regulatory risks across enterprise operations.
Sales Leader – Native-AI Practice
Enterprise sales leader driving net-new OpenAI consulting and implementation engagements for Blend, an AI services provider. Building North American pipeline, closing strategic accounts, and developing co-sell motions.
Gen AI Creative Technologist
Gen AI Creative Technologist building agentic workflows and creative systems for Pencil’s AI marketing platform. Enabling faster, scalable, on-brand advertising production.
Marketing Expert – Organic Growth, AI Agent Environments
Organic Growth expert designing and evaluating simulated SEO/AEO marketing tasks for Mercor. Creating realistic environments and rubrics to train and improve frontier AI models.
AI Session Annotator, French
French AI session annotator reviewing voice-and-camera assistant sessions for Mercor, which helps AI labs train frontier models. Scoring visual accuracy, language, timing, and safety warnings.
AI Session Annotator, English
AI session annotator evaluating English voice-and-camera assistant responses for Mercor’s AI training projects. Scoring visual accuracy, safety warnings, timing, and spatial-language use with evidence-based comments.
Marketing Expert – Lifecycle, AI Agent Environments
Lifecycle marketing expert designing realistic CRM and retention tasks for Mercor’s AI-agent training environments. Evaluating agent decisions, reference answers, and multi-tool workflows.
Generative AI Analyst, Traditional Chinese
Generative AI Analyst evaluating and annotating AI-generated content for Welo Data, a global AI data company. Reviewing multilingual data quality and providing structured feedback.
AI Systems, Zoho & Automation Specialist
AI automation specialist connecting Zoho, AI, CRM, WhatsApp, and data systems. Building reliable workflows for Panama real-estate brokerages.
AI Team Lead
AI Team Lead leading engineers and delivering automation, RPA, and LLM solutions. Driving AI strategy and adoption across Immense Group’s iGaming business.
Marketing Expert – Lifecycle, AI Agent Environments
Lifecycle marketing expert designing simulated environments to evaluate AI agents on real CRM and retention work. Creating briefs, rubrics, and reference answers for Mercor’s AI training projects.
Marketing Expert – Field Marketing, AI Agent Environments
Field marketing expert designing realistic task environments for Mercor’s AI training projects. Evaluating agent performance across events, pipeline recovery, sponsorships, and multi-tool workflows.
Senior Technical Director, AI/ML
AI/ML Senior Technical Director prioritizing cross-business AI initiatives for GE Electrification. Driving strategic roadmaps, risk management, and GenAI implementation with global teams.
Contract Trainer for Operations Management, Data & AI
Operations-Management-Honorartrainer:in für WBS TRAININGs digitale Weiterbildung. Coaching, Projektbewertung und praxisnaher Einsatz von Daten- und KI-Tools im Homeoffice.
Contract Trainer for Operations Management, Data & AI
Honorartrainer:in für Operations Management, Data und KI bei WBS TRAINING. Begleitung von Lernenden im Flipped-Classroom-Modell aus dem Homeoffice.
AI Enablement Engineer
AI Enablement Engineer building internal AI systems, agents, and integrations for Replicant's AI-powered customer-service platform. Enabling teams, managing AI foundations, and improving enterprise AI cost visibility.
Mid-Level Analyst/Developer, Java, Artificial Intelligence
Analista Java diagnosticando e documentando plataforma educacional interna com IA na Deliver IT. Mapeando requisitos, integrações e modernização sem desenvolvimento inicial.
AI Quality Evaluator, German
AI Quality Evaluator assessing German content, language quality, and cultural fit for Canva’s AI design experiences. Building datasets and refining labels with global localisation teams.
AI Safety Expert, English, Portuguese
AI safety experts red teaming conversational models for Mercor’s human-data AI projects. Probing vulnerabilities and generating reproducible datasets to improve frontier AI safety.
Software Evaluator – AI, IT, Data
Evaluator reviewing AI-generated software, IT, and data artifacts for Mercor. Grading quality, identifying errors, and providing structured feedback to train frontier AI systems.