An expert-level role focused on building the DVF AI platform capability by designing and implementing reusable AI platform services, APIs, libraries, templates and engineering patterns
3-year renewable Contract - Gauteng
Enable DVF AI use cases from discovery and prototype through production implementation
Engineer agentic and GenAI solutions including agents, RAG capabilities, prompts, context packages, tool integrations, evaluations and guardrails
Drive AI Native engineering by embedding AI into specification, development, testing, code review, documentation and operational workflows
Implement Snowflake capabilities and engineer data structures and integration patterns
Establish production AI engineering standards for evaluation, monitoring, observability, security and production-readiness
Ensure solutions address enterprise requirements for security, privacy, governance, information protection, auditability, resilience and operational support
Identify recurring requirements across use cases and convert them into shared DVF platform capabilities
Provide technical leadership through architecture and design discussions, code reviews, and mentoring engineers
Support production deployment, monitoring, incident resolution and continuous improvement
What they're looking for
Degree or diploma in Computer Science, Software Engineering, Artificial Intelligence, Machine Learning, Data Engineering, Information Technology, or a related discipline
Typically 7+ years of relevant engineering experience
Demonstrable experience building and operating complex production systems
Expert-level Python engineering capability
Demonstrable production experience with Generative AI/ LLM solutions
Strong data and integration engineering background
Practical experience implementing AI-assisted or AI-native engineering practices
Experience with Snowflake implementation or comparable enterprise cloud data platforms
Evidence of taking AI solutions beyond prototypes into governed, production-ready implementations
ESSENTIAL SKILLS:
AI Engineering
Expert knowledge of LLMs and Generative AI application engineering.
Hands-on experience building production-grade LLM applications and AI agents.
Strong understanding of Retrieval-Augmented Generation (RAG).
Prompt and context engineering.
Agent/tool integration and orchestration.
Structured outputs and tool calling.
Model and application evaluation.
AI guardrails and human approval mechanisms.
AI observability, reliability and production monitoring.
Responsible and secure AI engineering.
The source role already identifies agents, RAG, prompt/context engineering, tool integration, evaluation, guardrails and human approval gates as core AI engineering capabilities.
AI Native Engineering
Strong practical capability using AI throughout the software engineering lifecycle.
Specification-driven development.
AI-assisted implementation and refactoring.
AI-assisted test generation and quality engineering.
AI-supported code review and documentation.
Ability to build reusable AI-enabled developer tooling and workflows.
Strong understanding that AI-generated outputs still require accountable engineering review.
The reference specification explicitly expects responsible AI-assisted development across specification, coding, review, testing, documentation and operations while retaining accountability for correctness and security.
Software & Platform Engineering
Expert Python engineering.
Strong API and microservice engineering.
REST-based service integration.
Event-driven and asynchronous architecture.
Containers and Kubernetes.
CI/CD and automated testing.
Infrastructure as Code.
Secure software development.
Production observability and operational readiness.
Strong architecture and systems-thinking capability.
Python/Java, APIs, microservices, AWS/cloud-native engineering, Docker/Kubernetes/Terraform, automated testing, CI/CD and production readiness are all represented in the reference Expert specification.
Snowflake & Data Engineering
Strong Snowflake engineering and implementation capability.
Design and implementation of data models suitable for AI and application consumption.
Advanced SQL.
Data ingestion and transformation.
ETL/ELT engineering.
Metadata, lineage and data quality.
Secure data access patterns.
Integration between Snowflake, AI services, APIs and enterprise applications.
Performance and cost-conscious data engineering.
What's on offer
Flexible working of 1960 hours in a 12-month period