Client
Our client is Australia's national policing agency their role is to defend and protect Australia and Australia's future from domestic and global security threats.

Role
As the user researcher you will be responsible for design, organise, participate in and carry out user-centred research. You will develop a deep understanding of the users and advocate for recommendations that improve the user’s experiences with the service. You will use research findings to solve problems for the user and the team/Agency.

Responsibilities
Covert & Technical Operations (CTO) Transition Program has a requirement to engage a Senior Research Engineer to lead the design, development and operational evaluation of a secure human-in-the-loop artificial intelligence system for the triage of extremist and distressing digital material. This position is primarily an applied research and technical-delivery role. The successful candidate will work closely with researchers, investigators, technical specialists, wellbeing professionals and governance stakeholders to translate an operational problem into a rigorously evaluated prototype and partner-ready implementation pathway.

Technical leadership and development

Lead the technical design and implementation of a multimodal material-triage system incorporating relevant computer vision, language, audio and similarity-detection methods.
Develop capabilities for image and video-frame classification; extremist symbol and visual-marker detection; known-material hashing and near-duplicate detection; text and document classification; material-type and exposure-level categorisation; confidence calibration; and analyst review and correction.
Select, adapt and evaluate appropriate machine-learning models, prioritising reliable and explainable performance over unnecessary technical complexity.
Develop secure and reproducible data-processing, model-training and evaluation pipelines suitable for sensitive or restricted information.
Support deployment in on-premises, restricted-network or disconnected computing environments where required.

Operational translation

Work with partner personnel to document material-review workflows, sources of unnecessary exposure and operational requirements.
Translate investigator and wellbeing requirements into technical specifications, user stories, evaluation criteria and delivery milestones.
Develop or integrate a safe-review interface incorporating protected previews, blurred thumbnails, suppressed autoplay, representative-item selection and clear warning labels
Integrate prototype capabilities with relevant AiLECS platforms and approved partner systems.
Ensure that outputs retain appropriate provenance, model-version information, analyst decisions and audit records.

Evaluation and assurance

Establish representative development and held-out evaluation datasets in collaboration with authorised subject-matter experts.
Design and conduct rigorous evaluations addressing false-positive and false-negative performance, class imbalance, altered or obscured material, previously unseen content, model calibration, operational thresholds, investigator time, exposure reduction, usability and appropriate reliance.
Undertake adversarial and failure-mode testing and document known limitations, unsuitable uses and residual risks.
Contribute to privacy, security, ethical, legal and human-rights assessments associated with the project.
Work with organisational wellbeing specialists to ensure that exposure-reduction measures are appropriate and that evaluation does not introduce unnecessary exposure or inappropriate monitoring of individual employees.

Project and stakeholder management

Prepare and maintain a detailed technical plan, risk register, evaluation framework and delivery schedule.
Provide clear progress reporting to the AiLECS Director, project governance group and authorised partner representatives.
Coordinate technical contributions from research assistants, students, vendors and collaborating researchers where required.
Communicate technical findings and limitations clearly to specialist and non-specialist audiences.
Produce high-quality technical documentation, operational guidance, evaluation reports and deployment recommendations.

Research and capability development

Contribute to scholarly publications, technical reports, grant proposals and presentations arising from the project, subject to security, confidentiality and publication-approval requirements.
Identify opportunities to extend the capability to other forms of distressing digital evidence while maintaining appropriate safeguards and project scope.
Contribute to the wider research, translation and partnership objectives of the AiLECS Lab.

You will have

A postgraduate qualification in computer science, artificial intelligence, data science, software engineering or a related discipline, or an equivalent combination of relevant qualifications and substantial professional experience.
Demonstrated experience designing and delivering applied machine-learning systems beyond proof-of-concept stage.
Strong practical experience in at least two of the following: computer vision; image or video classification; natural-language processing; multimodal machine learning; similarity search or perceptual hashing; or audio analysis.
Advanced programming capability in Python and experience with contemporary machine-learning frameworks such as PyTorch.
Demonstrated experience building reproducible data pipelines, model-evaluation processes and deployable software components.
Strong understanding of classification performance, including class imbalance, precision, recall, false negatives, threshold selection, calibration and uncertainty.
Experience working with sensitive, confidential, regulated or security-controlled information.
Demonstrated ability to translate complex operational requirements into practical technical solutions.
Excellent stakeholder engagement and communication skills, including the ability to explain model behaviour, limitations and risks to non-technical decision-makers.
Demonstrated ability to manage a substantial technical project independently, work across organisational boundaries and deliver against agreed milestones.
Sound understanding of privacy, security, responsible AI, human oversight and the risks associated with high-consequence applications of machine learning.

Due to the nature of this role, this job is open to Australian Citizens only, you must hold a current NV1/NV2 Security clearance.

Calleo is an equal opportunity employer and we encourage applications from all people including Aboriginal and Torres Strait Islander peoples.

Follow Calleo on LinkedIn and visit our website to keep up to date on all our current job vacancies:
www.linkedin.com/company/calleoresourcing
www.calleo.com.au

Lead the technical design and implementation of a multimodal material-triage system incorporating relevant computer vision, language, audio and similarity-detection methods.
Develop capabilities for image and video-frame classification; extremist symbol and visual-marker detection; known-material hashing and near-duplicate detection; text and document classification; material-type and exposure-level categorisation; confidence calibration; and analyst review and correction.
Select, adapt and evaluate appropriate machine-learning models, prioritising reliable and explainable performance over unnecessary technical complexity.
Develop secure and reproducible data-processing, model-training and evaluation pipelines suitable for sensitive or restricted information.
Support deployment in on-premises, restricted-network or disconnected computing environments where required.

Operational translation

Work with partner personnel to document material-review workflows, sources of unnecessary exposure and operational requirements.
Translate investigator and wellbeing requirements into technical specifications, user stories, evaluation criteria and delivery milestones.
Develop or integrate a safe-review interface incorporating protected previews, blurred thumbnails, suppressed autoplay, representative-item selection and clear warning labels
Integrate prototype capabilities with relevant AiLECS platforms and approved partner systems.
Ensure that outputs retain appropriate provenance, model-version information, analyst decisions and audit records.

Evaluation and assurance

Establish representative development and held-out evaluation datasets in collaboration with authorised subject-matter experts.
Design and conduct rigorous evaluations addressing false-positive and false-negative performance, class imbalance, altered or obscured material, previously unseen content, model calibration, operational thresholds, investigator time, exposure reduction, usability and appropriate reliance.
Undertake adversarial and failure-mode testing and document known limitations, unsuitable uses and residual risks.
Contribute to privacy, security, ethical, legal and human-rights assessments associated with the project.
Work with organisational wellbeing specialists to ensure that exposure-reduction measures are appropriate and that evaluation does not introduce unnecessary exposure or inappropriate monitoring of individual employees.

Project and stakeholder management

Prepare and maintain a detailed technical plan, risk register, evaluation framework and delivery schedule.
Provide clear progress reporting to the AiLECS Director, project governance group and authorised partner representatives.
Coordinate technical contributions from research assistants, students, vendors and collaborating researchers where required.
Communicate technical findings and limitations clearly to specialist and non-specialist audiences.
Produce high-quality technical documentation, operational guidance, evaluation reports and deployment recommendations.

Research and capability development

Contribute to scholarly publications, technical reports, grant proposals and presentations arising from the project, subject to security, confidentiality and publication-approval requirements.
Identify opportunities to extend the capability to other forms of distressing digital evidence while maintaining appropriate safeguards and project scope.
Contribute to the wider research, translation and partnership objectives of the AiLECS Lab.

You will have

A postgraduate qualification in computer science, artificial intelligence, data science, software engineering or a related discipline, or an equivalent combination of relevant qualifications and substantial professional experience.
Demonstrated experience designing and delivering applied machine-learning systems beyond proof-of-concept stage.
Strong practical experience in at least two of the following: computer vision; image or video classification; natural-language processing; multimodal machine learning; similarity search or perceptual hashing; or audio analysis.
Advanced programming capability in Python and experience with contemporary machine-learning frameworks such as PyTorch.
Demonstrated experience building reproducible data pipelines, model-evaluation processes and deployable software components.
Strong understanding of classification performance, including class imbalance, precision, recall, false negatives, threshold selection, calibration and uncertainty.
Experience working with sensitive, confidential, regulated or security-controlled information.
Demonstrated ability to translate complex operational requirements into practical technical solutions.
Excellent stakeholder engagement and communication skills, including the ability to explain model behaviour, limitations and risks to non-technical decision-makers.
Demonstrated ability to manage a substantial technical project independently, work across organisational boundaries and deliver against agreed milestones.
Sound understanding of privacy, security, responsible AI, human oversight and the risks associated with high-consequence applications of machine learning.

Lead the technical design and implementation of a multimodal material-triage system incorporating relevant computer vision, language, audio and similarity-detection methods.
Develop capabilities for image and video-frame classification; extremist symbol and visual-marker detection; known-material hashing and near-duplicate detection; text and document classification; material-type and exposure-level categorisation; confidence calibration; and analyst review and correction.
Select, adapt and evaluate appropriate machine-learning models, prioritising reliable and explainable performance over unnecessary technical complexity.
Develop secure and reproducible data-processing, model-training and evaluation pipelines suitable for sensitive or restricted information.
Support deployment in on-premises, restricted-network or disconnected computing environments where required.

Operational translation

Work with partner personnel to document material-review workflows, sources of unnecessary exposure and operational requirements.
Translate investigator and wellbeing requirements into technical specifications, user stories, evaluation criteria and delivery milestones.
Develop or integrate a safe-review interface incorporating protected previews, blurred thumbnails, suppressed autoplay, representative-item selection and clear warning labels
Integrate prototype capabilities with relevant AiLECS platforms and approved partner systems.
Ensure that outputs retain appropriate provenance, model-version information, analyst decisions and audit records.

Evaluation and assurance

Establish representative development and held-out evaluation datasets in collaboration with authorised subject-matter experts.
Design and conduct rigorous evaluations addressing false-positive and false-negative performance, class imbalance, altered or obscured material, previously unseen content, model calibration, operational thresholds, investigator time, exposure reduction, usability and appropriate reliance.
Undertake adversarial and failure-mode testing and document known limitations, unsuitable uses and residual risks.
Contribute to privacy, security, ethical, legal and human-rights assessments associated with the project.
Work with organisational wellbeing specialists to ensure that exposure-reduction measures are appropriate and that evaluation does not introduce unnecessary exposure or inappropriate monitoring of individual employees.

Project and stakeholder management

Prepare and maintain a detailed technical plan, risk register, evaluation framework and delivery schedule.
Provide clear progress reporting to the AiLECS Director, project governance group and authorised partner representatives.
Coordinate technical contributions from research assistants, students, vendors and collaborating researchers where required.
Communicate technical findings and limitations clearly to specialist and non-specialist audiences.
Produce high-quality technical documentation, operational guidance, evaluation reports and deployment recommendations.

Research and capability development

Contribute to scholarly publications, technical reports, grant proposals and presentations arising from the project, subject to security, confidentiality and publication-approval requirements.
Identify opportunities to extend the capability to other forms of distressing digital evidence while maintaining appropriate safeguards and project scope.
Contribute to the wider research, translation and partnership objectives of the AiLECS Lab.

You will have

A postgraduate qualification in computer science, artificial intelligence, data science, software engineering or a related discipline, or an equivalent combination of relevant qualifications and substantial professional experience.
Demonstrated experience designing and delivering applied machine-learning systems beyond proof-of-concept stage.
Strong practical experience in at least two of the following: computer vision; image or video classification; natural-language processing; multimodal machine learning; similarity search or perceptual hashing; or audio analysis.
Advanced programming capability in Python and experience with contemporary machine-learning frameworks such as PyTorch.
Demonstrated experience building reproducible data pipelines, model-evaluation processes and deployable software components.
Strong understanding of classification performance, including class imbalance, precision, recall, false negatives, threshold selection, calibration and uncertainty.
Experience working with sensitive, confidential, regulated or security-controlled information.
Demonstrated ability to translate complex operational requirements into practical technical solutions.
Excellent stakeholder engagement and communication skills, including the ability to explain model behaviour, limitations and risks to non-technical decision-makers.
Demonstrated ability to manage a substantial technical project independently, work across organisational boundaries and deliver against agreed milestones.
Sound understanding of privacy, security, responsible AI, human oversight and the risks associated with high-consequence applications of machine learning.

Lead User Researcher

Share this job

Share this job

Similar Jobs

Similar Jobs