AI Engineer (Full Stack)

Business Unit: IT & Digital Implementation

Role Type: Contract

Location: Centurion

Role Purpose To design, build, and operationalise AI-powered solutions and full-stack applications that improve operational efficiency across the organisation. The role focuses on fine-tuning LLMs, implementing RAG architectures, building agentic workflows, and delivering copilots/automation, while also engineering reliable backend and frontend systems integrated into enterprise platforms and processes.

Responsibilities and work outputs Process

• Define and deliver an AI Full Stack delivery roadmap aligned to business objectives and operational efficiency goals.
• Translate business problems into AI solution designs, including RAG patterns, agentic orchestration, and automation workflows.
• Build and maintain LLM fine-tuning pipelines (SFT/LoRA/PEFT where applicable), including dataset preparation, evaluation, safety checks, and regression testing.
• Design and implement RAG solutions using multiple patterns (e.g., basic RAG, hybrid search, multi-vector, parent-child chunking, query rewriting, reranking, self-reflection, contextual retrieval).
• Architect and implement agentic systems (tool use, planning, memory, guardrails) including Hierarchical Agents and Multi-Agent collaboration models.
• Develop AI copilots for operational use cases (support, incident triage, knowledge retrieval, automation execution, reporting).
• Implement AI automation workflows integrating LLMs with enterprise tools and APIs (ticketing, monitoring, workflow engines, data platforms).
• Build visual text extraction solutions using V-LLMs / multimodal models for complex documents, including handwritten content, scanned PDFs, tables, forms, and mixed layouts.
• Engineer robust backend services (Java/Spring Boot and Python) exposing APIs for AI services, orchestration, document processing, and data access.
• Engineer frontend experiences (React and/or Angular) for copilots, admin consoles, evaluation dashboards, and workflow user interfaces.
• Implement and maintain event-driven and integration patterns using Apache Camel and messaging (Kafka

/ ActiveMQ / RabbitMQ).

• Contribute to BPMN engineering and workflow automation (e.g., Camunda), integrating AI into process execution and decision support.
• Apply software engineering best practices: secure coding, code reviews, CI/CD, observability, testing (unit/integration), and documentation.
• Establish model and system monitoring (latency, cost, quality, drift, retrieval relevance, hallucination rate) and continuous improvement loops.
• Ensure solutions align to enterprise architecture standards, data governance, security, and compliance requirements.

Client

• Build strong relationships with internal stakeholders and delivery teams (Ops, Product, Security, Data, Architecture).
• Gather requirements and shape them into prioritised backlogs and technical delivery plans.
• Communicate progress, risks, and dependencies clearly; manage expectations through demos, workshops, and show-and-tells.
• Deliver against agreed service levels for reliability, performance, and supportability of AI-enabled services.
• Provide recommendations to improve user experience, fairness, and responsible AI usage within the area of responsibility.

People

• Contribute to a positive engineering culture that encourages experimentation, learning, and delivery excellence.
• Share knowledge through documentation, walkthroughs, and enabling others to operate and extend AI solutions.
• Collaborate effectively in cross-functional squads (developers, architects, analysts, product owners, QA, DevOps).
• Mentor junior team members where applicable and promote consistent engineering standards.

Finance

• Deliver solutions with awareness of cost drivers (LLM usage, inference, embedding, vector storage, compute).
• Optimise architecture for cost/performance (caching, batching, reranking strategy, model selection, token efficiency).
• Support business cases with measurable benefits (time saved, reduced incidents, improved throughput, improved accuracy).
• Ensure tooling and implementation decisions align to budget constraints and procurement standards where required.

Knowledge

• Strong understanding of LLMs, prompt engineering, fine-tuning approaches (SFT/LoRA/PEFT), and evaluation techniques.
• Deep knowledge of RAG and information retrieval: embeddings, chunking strategies, hybrid search, reranking, metadata filtering, and relevance testing.
• Experience designing agentic workflows: tool/function calling, planning, memory, guardrails, hierarchical and multi-agent systems.
• Knowledge of multimodal/V-LLM document understanding: layout parsing, table extraction, handwriting recognition approaches, and document QA.
• Strong software engineering fundamentals and system design for scalable services.
• Proficiency in:
• Backend: Java, Spring Boot, Python
• Frontend: React, Angular
• Databases: PostgreSQL, unstructured databases (e.g., document stores), vector databases/search platforms
• Integration: Apache Camel, Kafka / ActiveMQ / RabbitMQ
• Workflow: BPMN engineering, workflow orchestration (e.g., Camunda)
• CI/CD, observability, testing practices, and security-by-design.
• Understanding of governance, risk, compliance, and secure handling of sensitive data in AI systems.

Competencies

• Making Decisions
• Managing Tasks
• Generating Ideas
• Taking Action
• Providing Insights
• Empowering Individuals
• Directing People (as needed within delivery leadership)
• Systems Thinking & Problem Solving
• Stakeholder Management & Communication
• Experimentation Discipline (hypothesis → test → measure → improve)
• Engineering Quality Mindset (reliability, security, maintainability)

Experience and Qualifications

• 2 – 5 years experience in software engineering (full stack and/or backend) with demonstrated delivery of production systems.
• Hands-on experience implementing at least some of: LLM fine-tuning, RAG solutions, agentic workflows, AI automation, or copilot-style applications.
• Experience integrating systems using messaging and integration frameworks (Kafka/ActiveMQ/RabbitMQ, Apache Camel).
• Experience working with structured and unstructured data, including document-heavy workflows.
• Exposure to Agile delivery and modern engineering practices (CI/CD, testing, monitoring).
• Bachelor’s degree in Computer Science, Engineering, Data Science (or equivalent practical experience).
• Advantageous: certifications or proven experience in ML/AI engineering, cloud platforms, or DevOps/MLOps.