AI Safety Engineer

AI Safety Engineer (Multiple Opportunities) - AI Safety Institute, Science & Technical Level 7-9, Technology & Digital Join the Department of Industry, Science and Resources as an AI Safety Engineer with the new Australian AI Safety Institute (AISI). This role focuses on research, policy advice, and technical evaluation of frontier AI systems. Locations: Canberra, Sydney, Darwin, Brisbane, Adelaide, Hobart, Melbourne, Perth Application closing date: 18 January 2026 Salary range: $122,235 – $172,828 About The Department The Department of Industry, Science and Resources drives Australia’s economic agenda through science and technology. It builds a resilient, productive and sustainable economy for all Australians. About The AI Safety Institute The Australian AI Safety Institute (AISI) will be the government’s hub of AI safety expertise, operating with transparency, responsiveness and technical rigor. AISI will conduct technical assessments, support coordinated government action, foster international engagement, and publish research to inform industry, academia and the public. The Division Part of the Technology and Digital Policy Division, the AISI provides policy advice, program delivery and engagement on enabling and critical technologies as well as the digitisation of the economy. Opportunity As a founding team member, you will help shape Australia’s monitoring, testing and governance of AI. You will assess risks from frontier models, manage safety-critical evaluations, and collaborate with domestic and international experts on emerging AI safety standards. Ideal Candidate – Senior AI Safety Engineer (Pay Scale 8 and 9) Extensive hands‑on experience with frontier or near‑frontier AI models (LLMs, multimodal systems, agentic frameworks). Experience building and running evaluations of frontier AI systems or safety‑relevant behaviours. Experience creating or using safety‑tooling such as red‑team frameworks, test harnesses, automated pipelines, or monitoring systems. Progressive implementation and stress‑testing safeguards: guardrails, filtering systems, access controls, safety‑tuning, inference‑time controls. Large‑scale behavioural evaluation experience, including data sets, diagnostics and debugging. Knowledge of safety‑relevant AI failure modes: robustness, jailbreaks, unintended behaviours, reliability. Strong collaboration with research scientists and engineers; multidisciplinary teamwork. 5+ years of industry or academic experience on training, tuning, evaluating or operating advanced AI models. Ideal Candidate – AI Safety Engineer (Pay Scale 7 and 8) Hands‑on experience with frontier or near‑frontier AI models and systems. Support or contribute to evaluations of frontier AI systems or safety‑relevant behaviours. Use safety‑tooling such as red‑team frameworks, test harnesses, automated pipelines, or monitoring systems. Implement or test safety mitigations: guardrails, filtering, access controls, safety‑tuning, inference‑time controls. Contribution to large‑scale behavioural evaluations: logs, datasets, debugging. Understanding of common safety‑relevant AI failure modes. Effective in multidisciplinary teams; contribution to operational delivery of evaluation work. 3+ years of industry or academic experience on training, tuning, evaluating or operating advanced AI models. What You Will Do As a Senior AI Safety Engineer Operationalise evaluation designs, translating research concepts into practical, reproducible experiments. Build, maintain and operate evaluation and safety‑testing tooling for frontier AI systems. Run large‑scale behavioural tests, produce empirical evidence for safety analysis. Diagnose emerging failure modes; identify vulnerabilities and anomalous behaviours. Document evaluation logs, testing reports and safeguard assessments. Improve engineering practices, tooling and testing infrastructure. Collaborate across government, industry and academia, participate in international AI safety initiatives. Contribute to technical reports and research outputs; shape the culture and reputation of the AISI. As an AI Safety Engineer Support the implementation of evaluation designs and translate testing methodologies into repeatable experiments. Support operation and maintenance of evaluation and safety‑testing tooling. Assist in running behavioural tests and generating reliable evidence for safety analysis. Identify emerging failure modes and collaborate with scientists to assess risks. Maintain evaluation logs, testing reports, safeguard assessments. Contribute to improvements in engineering practices, tooling and infrastructure. Collaborate across sectors and participate in international AI safety initiatives. Contribute to technical reports and research outputs; shape AISI culture. Eligibility Ability to obtain baseline security clearance and, if required, higher clearance. Australian citizenship required for employment in the APS. Notes The successful candidate will be signed to an initial 12‑month contract with the possibility of extension. Salaries range from $122,235 for a Research Engineer (Level 7) to $172,828 for a Senior Research Engineer (Level 9), corresponding to APS Executive Levels 1 and 2. Please provide a 2‑page CV (or up to 4 pages if including publications). A bachelor’s degree in a relevant field is highly desirable. Merit pool may be established to fill future vacancies within 18 months of advertisement. Flexible work arrangements are available; this vacancy applies Australia‑wide and remote work may be considered. Application Information Your application must not contain any classified or sensitive information. The selection panel will not consider applications containing classified content. Please provide a pitch (≤750 words) explaining how your skills, knowledge and experience are relevant to this role and why you are the best candidate. Complete your application online and upload your CV in .doc, .docx or .pdf format. Accessible documentation is available on request. Contact or call if you need assistance. For further application details, refer to the department's “Applying for a position” information. Contact Information Bill Black – – Seniority Level Mid‑Senior level Employment Type Contract Job Function Management and Manufacturing #J-18808-Ljbffr