Lead Data Scientist - Financial Crime at Westpac Group in Australia

Lead Data Scientist - Financial Crime

Location: Australia

• Sydney, Melbourne, Brisbane or Perth Location with Hybrid Working.
• Set the technical direction for fraud and financial-crime data science.
• Assure consequential decisions, develop senior practitioners and turn portfolio investment into controlled outcomes.

What’s the role?

As a Lead Data Scientist, you will be the senior technical authority for fraud and financial-crime data science within Westpac’s Enterprise Functions squad. You will set direction across a portfolio of detection, monitoring and investigation capabilities, ensuring outcomes are valid, explainable, operationally useful and well governed.

The portfolio spans transaction monitoring, suspicious matter investigations, scams, anti-money laundering and counter-terrorism financing (AML/CTF), entity resolution, typology and network analysis, anomaly detection, natural language processing (NLP), Generative AI, neural networks and graph neural networks (GNNs). You will resolve the hardest methodological and architectural trade-offs and assure material work.

This is a lead individual contributor and craft-leadership role (AA Lead Specialist), not automatically a line-management position. You will operate through technical authority, standards and influence, delegate technical work and remain accountable for coherence and quality across initiatives. You will develop senior practitioners and represent the domain with senior business, technology and risk stakeholders.

The impact you can make

• Shape a coherent portfolio across prevention, detection, prioritisation, investigation and continuous control improvement to protect customers and the financial system.
• Direct investment towards measurable customer and control outcomes, balancing investigative need, data readiness, feasibility, regulatory risk and learning value.
• Improve portfolio quality through common standards, independent challenge and comparable evaluation, rather than isolated analytical experiments.
• Build enduring technical capability, develop senior leaders and reduce dependence on individual knowledge holders.

What you’ll be doing

• Define and maintain the domain data science roadmap. Anticipate changes in criminal behavioural, scams, channels, regulation, data and technology, and turn them into evidence-based priorities.
• Establish reference approaches combining rules, typologies, statistical detection, classical machine learning, graph analytics, neural networks, GNNs, NLP and GenAI. Set principles for choosing simpler, more explainable methods and common entity, event, graph, feature, label and decision-policy definitions.
• Resolve consequential choices involving anomaly detection, weak or delayed labels, entity resolution, graph construction, temporal validation, thresholds, causal claims and human decisions. Lead or sponsor dynamic graphs, embeddings, community and path analysis, GNNs and hybrid rules-plus-ML systems.
• Set the direction for investigation summarisation, information extraction, narrative support, typology discovery and investigator copilots. Define GenAI and agent evaluation for grounding, factuality, completeness, consistency, safety, human review, prompt injection and data-exfiltration risk.
• Connect model outputs to investigator queues, escalation, deferral, feedback and approved human decision rights. Sponsor champion–challenger approaches and independent challenge for material models, typologies and GenAI applications.
• Review high-risk models, graphs, prompts, agents, pipelines and decision policies before material release or change. Set standards for data quality, leakage prevention, temporal validation, calibration, rare-event metrics, fairness, explainability, reproducibility and documentation.
• Establish comparable evaluation linking technical performance to alert yield, losses prevented, investigation effort, customer impact, missed risk and control effectiveness. Ensure monitoring covers drift, graph and entity changes, typology decay, fairness, operational load and GenAI quality.
• Define triggers and governance for recalibration, retraining, prompt or typology changes, rollback, suspension and retirement. Lead the technical response to material model or data incidents and embed systemic lessons into standards and controls.
• Lead Responsible AI, Model Risk, privacy, security, AML/CTF, records-management and audit assurance. Act as senior technical counterpart to legal, compliance and second-line partners; provide credible governance, regulatory and audit evidence within delegated authority.
• Lead difficult fairness, proportionality, vulnerable-customer and human-accountability decisions, escalating beyond delegated authority. Preserve the distinction between indicators, model inference and verified evidence; test sensitive-data use, consequential failure and misuse under approved controls.
• Partner with financial-crime leadership, investigators and product owners to align strategic requirements, propose portfolio-level options and guide investment through delivery, adoption and outcome realisation. Explain uncertainty, scenarios, benefits and control trade-offs to senior stakeholders.
• Resolve dependencies across data platforms, case-management systems, entity services, cloud/AI platforms and control owners. Integrate solutions into sustainable operations, and redirect or stop work where value is limited or risk unacceptable.
• Create reusable reference implementations, feature and graph patterns, typology components, evaluation suites, model cards and design guidance. Set review standards, coach Senior Data Scientists towards Lead-level authority and build communities across science, investigation, engineering and assurance.
• Lead targeted horizon scanning and experimentation in adaptive anomaly detection, multimodal and agentic AI, temporal graphs and GNNs. Adopt new methods only with evidence of material advantage and represent the squad and craft in senior technical, risk and professional forums.

What do I need?

• Deep, sustained expertise in advanced analytics, statistics and machine learning, with evidence of technical leadership across multiple consequential production use cases.
• Advanced-to-Mastery depth in at least one fraud or financial-crime analytical specialism, with Advanced breadth across several areas: anomaly and behavioural detection; temporal network science, graph embeddings and GNNs; entity resolution; NLP, information retrieval and GenAI; neural networks and sequence modelling; or decision policy and human-in-the-loop design.
• Expert command of rare-event evaluation, temporal validation, uncertainty, calibration, model and graph explainability, champion–challenger testing and translating performance into decisions. Ability to assess interactions among data, models, rules, prompts, investigators, capacity and controls.
• Strong working knowledge of production AI architecture, MLOps/LLMOps, monitoring, observability, controlled release, reliability and lifecycle governance. A record of establishing standards, assuring others’ work and resolving methodological disputes with evidence.
• Deep understanding of fraud and financial-crime detection and investigation, including transaction monitoring, scams, AML/CTF, suspicious matter investigations, typologies, red flags, entity risk and network behavioural.
• Extensive experience leading fraud and financial-crime solutions or portfolios in a large, complex organisation at a scale comparable to Westpac, such as major financial services, telecommunications or a similarly regulated enterprise.
• Strong understanding of criminal adaptation, data and label limitations, investigator workflows, false-positive burden, customer friction, vulnerable-customer impacts and operational control design.
• Advanced understanding of Responsible AI, Model Risk, privacy, AI security, auditability, AML/CTF and AUSTRAC expectations. Demonstrated capability to represent Westpac with external regulators, within delegated authority, explaining methods, evidence, limitations, controls and remediation.
• Experience influencing senior business, investigation, technology and risk stakeholders on material technical and portfolio decisions. Ability to define a roadmap, evaluate build/buy/partner options and prioritise competing investments against measurable outcomes.
• Enterprise-minded judgement, technical courage and intellectual honesty. Make difficult calls, expose uncertainty and dissent, remain calm under consequence and balance innovation with proportionality, customer rights and regulatory obligations.
• The ability to connect strategy to implementation detail, influence without hierarchy and develop senior technical leaders. You need breadth to assure integrated solutions and recognised depth for consequential decisions, not personal implementation of every specialist component.

What success looks like

• The portfolio delivers adopted, measurable and sustainable detection, monitoring and investigation capabilities, with clear evidence of customer and control outcomes.
• Consequential analytical decisions withstand independent challenge, and senior stakeholders understand the roadmap, uncertainty, trade-offs and risks.
• Quality improves across initiatives through common definitions, reusable patterns, disciplined review, monitoring and timely intervention.
• Senior practitioners grow in technical authority, knowledge is shared and the domain is less dependent on individual specialists.
• Decision rights remain clear: you own analytical direction, method quality and model assurance, not accountable business decisions, authorised investigations, compliance, second-line risk or model-agnostic cloud, security and platform infrastructure.

Ready to build what matters?

Apply now and show us what you have built, how you approached the problem, and what changed because your solution made it into the hands of users!

To get started, simply click on the APPLY or APPLY NOW button. Please note that application closing dates are subject to change so don’t delay your application!

We’re all about creating a supportive and inclusive community. We welcome everyone – no matter your age, gender, background, or abilities. We also provide additional support to welcome our veterans, Indigenous Australians, and neurodiverse community.

If you need any adjustments during the recruitment process, you can find out more information and additional contact details by visiting the "People with Disability and/or needing Accessibility Requirements" page on our website.