The LHR Group
Website:
thelhr.group
Job details:
Role Overview: As the AI Safety & Evaluation Lead, you will ensure AI systems are accurate, safe, unbiased, secure, and compliant. With students as the primary users, you will focus on student psychology, content safety, and handling sensitive queries in an age-appropriate manner. This is a 0-to-1 role requiring high ownership, hands-on execution, adaptability, and comfort with ambiguity and fast-paced work.
Key Responsibilities
- Own AI evaluation frameworks, including automated, human-in-the-loop, and benchmark testing.
- Lead red-teaming, adversarial testing, and vulnerability identification across AI models.
- Define AI safety, responsible AI, bias, fairness, and child-safe content standards.
- Build and maintain evaluation tools using platforms such as RAGAS, LangSmith, PromptBench, and LMEval.
- Conduct bias, fairness, hallucination, grounding, and response-quality assessments across languages and regions.
- Manage AI risk assessments and translate findings into actionable model or prompt improvements.
- Establish AI governance documentation, audit trails, and safety reports.
- Embed evaluation and safety checks into AI release pipelines with engineering teams.
- Track relevant AI governance and regulatory frameworks, including NIST AI RMF and India/EU AI regulations.
Requirements
- Strong experience in AI/LLM safety, evaluation, and risk assessment.
- Expertise in model evaluation, red-teaming, and adversarial testing.
- Strong understanding of bias, fairness, hallucination, grounding, and content safety.
- Experience working with production AI/LLM systems and evaluation tools.
- Knowledge of AI governance, compliance, and responsible AI practices.
- Understanding of student psychology and age-appropriate AI output design.
Click on Apply to know more.