LLM Safety Evaluator

Role Overview: As an AI Safety Red Teamer proficient in both English and Telugu, your primary responsibility will be to assess conversational models and agents through various techniques such as jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation. You will be tasked with generating high-quality annotated data by classifying vulnerabilities, flagging failures, and identifying systemic risks. Additionally, you will apply structured frameworks, taxonomies, and benchmarks to ensure consistent and reproducible testing. Your role will also involve producing detailed reports, datasets, and documented attack cases that stakeholders can effectively act upon. Key Responsibilities: - Red team conversational models and agents using jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation - Generate high-quality annotated data by classifying vulnerabilities, flagging failures, and identifying systemic risks - Apply structured frameworks, taxonomies, and benchmarks for consistent and reproducible testing - Produce detailed reports, datasets, and documented attack cases for stakeholder action Qualifications Required: - Strong experience in red teaming, adversarial probing, or cybersecurity (penetration testing, exploit development, or socio-technical risk analysis) - Strong experience in working with LLMs or conversational systems, including prompt injection and adversarial input design - Strong experience in communicating technical risk clearly to both technical and non-technical audiences - Strong experience in applying structured methodologies rather than ad hoc testing approaches - Native fluency in both English and Telugu is required Role Overview: As an AI Safety Red Teamer proficient in both English and Telugu, your primary responsibility will be to assess conversational models and agents through various techniques such as jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation. You will be tasked with generating high-quality annotated data by classifying vulnerabilities, flagging failures, and identifying systemic risks. Additionally, you will apply structured frameworks, taxonomies, and benchmarks to ensure consistent and reproducible testing. Your role will also involve producing detailed reports, datasets, and documented attack cases that stakeholders can effectively act upon. Key Responsibilities: - Red team conversational models and agents using jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation - Generate high-quality annotated data by classifying vulnerabilities, flagging failures, and identifying systemic risks - Apply structured frameworks, taxonomies, and benchmarks for consistent and reproducible testing - Produce detailed reports, datasets, and documented attack cases for stakeholder action Qualifications Required: - Strong experience in red teaming, adversarial probing, or cybersecurity (penetration testing, exploit development, or socio-technical risk analysis) - Strong experience in working with LLMs or conversational systems, including prompt injection and adversarial input design - Strong experience in communicating technical risk clearly to both technical and non-technical audiences - Strong experience in applying structured methodologies rather than ad hoc testing approaches - Native fluency in both English and Telugu is required