Senior Graph Data Scientist / TigerGraph Platform Engineer

Role purpose

To lead the design, engineering, deployment and operation of enterprise-scale graph data platforms and graph-based analytics solutions across the bank.

The role combines deep expertise in TigerGraph, graph analytics, graph machine learning and knowledge graphswith hands-on ownership of AKS Kubernetes-based infrastructure, cloud-native deployment, data ingestion, performance optimisation and operational resilience.

The successful candidate will enable relationship-driven intelligence across priority use cases including fraud detection, financial crime, AML, customer intelligence, network risk management, AI and GenAI.

Key responsibilities

Graph platform architecture and engineering

• Architect, design, deploy and operate secure, scalable and highly available TigerGraph clusters on Azure Kubernetes Service (AKS).

• Build and manage distributed graph infrastructure, including containerisation, orchestration, autoscaling, workload isolation, cluster management, monitoring and fault tolerance.

• Configure and optimise networking, storage, compute, identity, access control, secrets management and security controls for graph workloads in an enterprise environment.

• Ensure high availability, platform resilience, performance, capacity management and cost efficiency across TigerGraph and Kubernetes environments.

• Evaluate emerging graph, Kubernetes and cloud technologies to inform platform evolution and roadmap decisions.

Graph data modelling and analytics

• Lead the design of advanced graph data models that represent complex relationships across customers, accounts, transactions, devices, merchants, organisations and other enterprise entities.

• Develop high-performing GSQL queries, graph algorithms and analytical engines to uncover hidden relationships, suspicious networks, behavioural patterns and business insights.

• Apply graph techniques including community detection, link prediction, path analysis, centrality, similarity analysis, entity resolution and network-risk scoring.

• Optimise graph query performance, workload throughput and resource utilisation across large-scale distributed graph environments.

• Design reusable graph-derived features to enhance downstream machine-learning models, decisioning systems and risk-scoring capabilities.

Graph machine learning, AI and knowledge graphs

• Develop and operationalise graph-based machine-learning solutions, including graph neural networks and relationship-aware predictive models.

• Build and manage enterprise knowledge graphs that support advanced analytics, semantic intelligence, GenAI and retrieval-augmented generation use cases.

• Enable graph-enhanced AI solutions by connecting structured and unstructured enterprise information through relationship-centric data models.

• Partner with data scientists, AI engineers and business teams to translate graph insights into measurable business outcomes.

• Monitor and improve model accuracy, feature effectiveness, model performance and operational outcomes over time.

Data integration and operationalisation

• Design and implement secure, scalable data-ingestion pipelines into TigerGraph from enterprise platforms such as Azure Data Lake Storage, Databricks, APIs, transactional systems and streaming data sources.

• Support both batch and real-time graph data ingestion, transformation and enrichment processes.

• Ensure graph solutions integrate effectively with enterprise data platforms, APIs, data products, risk systems and decisioning engines.

• Develop CI/CD pipelines for graph applications, infrastructure and GSQL assets using Kubernetes-native and DevOps tooling.

• Establish monitoring, alerting, logging, observability and incident-management practices for graph platforms and graph-based services.

Financial crime and enterprise use cases

• Deliver graph analytics solutions for fraud detection, financial crime, AML, suspicious-network identification, customer intelligence and network-risk management.

• Translate highly connected and complex financial-services data into practical, explainable and actionable business solutions.

• Support risk, fraud, compliance, customer and AI teams in identifying, prioritising and delivering high-value graph use cases.

• Ensure solutions meet enterprise requirements for security, governance, privacy, auditability and resilience.

Leadership and stakeholder engagement

• Provide technical leadership and thought leadership on graph analytics, graph ML, TigerGraph, Kubernetes and graph-driven AI strategy.

• Mentor engineers and data scientists on graph data modelling, GSQL development, graph algorithms, Kubernetes operations and graph-based ML techniques.

• Communicate complex graph, infrastructure and AI concepts clearly to both technical and business stakeholders.

• Champion experimentation, innovation, reusable engineering standards and best practice across graph and distributed-systems capabilities.

• Support strategic digital, data and AI transformation programmes by embedding scalable graph capabilities into Nedbank’s enterprise ecosystem.

Required experience

• 7+ years’ experience in data engineering, graph engineering, data science, platform engineering, cloud architecture or related technical roles.

• Proven hands-on experience with TigerGraph, including graph data modelling, GSQL development, loading jobs, query optimisation and graph analytics.

• Experience architecting, deploying and operating TigerGraph or comparable distributed graph platforms in production environments.

• Strong practical experience with Azure Kubernetes Service (AKS), Kubernetes cluster operations, containerisation, orchestration, autoscaling and workload management.

• Experience with distributed systems, cloud-native architecture, microservices and enterprise platform integration.

• Strong Python capability for graph analytics, data engineering, machine learning and automation.

• Experience implementing graph algorithms such as community detection, centrality, link prediction, similarity analysis, path analysis and entity resolution.

• Experience with graph-based machine learning, graph neural networks or graph-derived feature engineering.

• Experience integrating graph platforms with ADLS, Databricks, APIs, enterprise data platforms and streaming or near-real-time data pipelines.

• Experience building CI/CD pipelines, infrastructure-as-code, observability and production monitoring for cloud-native platforms.

• Strong SQL, data-modelling and data-engineering expertise.

• Experience in financial services, fraud, AML, financial crime, risk analytics or complex customer-network analysis.

Preferred experience

• TigerGraph certification.

• Certified Kubernetes Administrator (CKA) or Certified Kubernetes Application Developer (CKAD).

• Azure certifications, particularly Azure Kubernetes Service, Azure Data Engineering or Azure Architecture certifications.

• Experience with Azure Databricks, Azure Data Lake Storage, Azure DevOps, Terraform, Helm, Docker and GitOps practices.

• Experience with Kafka, event streaming or real-time data-processing platforms.

• Knowledge-graph, semantic-model or ontology experience.

• Experience with GenAI, RAG, agentic AI or graph-enhanced AI applications.

• Exposure to fraud detection, AML transaction monitoring, sanctions screening, KYC, entity resolution or suspicious-network analysis.

Qualifications

Essential

• Bachelor’s degree in Computer Science, Engineering, Mathematics, Statistics, Data Science or another relevant STEM discipline.

Preferred

• Master’s degree in Computer Science, Engineering, Mathematics, Statistics, Data Science or a related field.

• Relevant TigerGraph, Kubernetes and Azure cloud certifications.

Key competencies

• Deep technical curiosity and strong problem-solving ability.

• Ability to design scalable, resilient and secure enterprise-grade platforms.

• Strong ownership mindset, with the ability to move from architecture through implementation and operational support.

• Ability to translate complex relationship-based data into commercially valuable outcomes.

• Strong communication skills across technical, business, risk and executive stakeholders.

• High standards for engineering quality, performance, governance and operational discipline.

• Comfortable working in a complex, regulated financial-services environment.

Measures of success

• A scalable, secure and highly available TigerGraph platform operating successfully on AKS.

• High-performing graph ingestion pipelines and GSQL workloads integrated with enterprise data platforms.

• Production-ready graph analytics and graph ML solutions delivered for fraud, financial-crime and customer-intelligence use cases.

• Measurable improvement in fraud detection, suspicious-network identification, risk insight, false-positive reduction or decisioning effectiveness.

• Reusable graph data models, graph-derived features, knowledge graphs and engineering standards established.

• Effective platform monitoring, automation, resilience, cost optimisation and operational support in place.