Lead AI Engineer (LLMs & Agents)
HybridAtholl Gardens, Gauteng, South Africa
Job Summary
Architect, design, and develop advanced AI systems focusing on LLMs and agentic solutions, including multi-agent orchestration, tool design, and memory-augmented workflows. Lead teams, mentor junior engineers, and collaborate with cross-functional experts to deliver AI-driven innovations for global clients in banking, insurance, healthcare, and telecommunications. Own the entire AI development lifecycle from research and prototyping to deployment, monitoring, and continuous improvement through production feedback loops. Develop and maintain production-ready AI solutions ensuring scalability, performance, reliability, and robust observability via tracing and evaluation frameworks. Deliver tangible outcomes such as underwriting agents, knowledge agents, and workflow optimizers while evaluating frontier models from providers like Anthropic, OpenAI, Google, and Meta.
Required Qualifications
- Deep experience with LLMs and Agents
- Hands-on experience building and deploying production agentic systems
- Experience with multi-agent architectures
- Experience with tool/function calling design
- Experience with agent memory models
- Familiarity with orchestration frameworks such as LangChain, Google ADK, or AutoGen
- Working knowledge of open standards like the Model Context Protocol (MCP)
- Solid background in programming
- Past experience in machine learning
- Deep understanding of at least one area such as deep learning, reinforcement learning, fine-tuning, or model development
- Ability to evaluate and select across frontier and open-source models from providers including Anthropic, OpenAI, Google, and Meta
- Ability to articulate the tradeoffs between hosted inference, self-hosted deployment, fine-tuning, RAG, and agentic routing for a given use case
- Excellent problem-solving, analytical, and critical thinking skills
- Proficiency in applying mathematical principles to analyze, model, and solve complex problems
- Experience designing LLM evaluation frameworks
- Experience implementing tracing and observability tooling (e.g. Langfuse, LangSmith)
- Experience building feedback loops that drive continuous improvement from production data
- Experience working in cross-functional teams
- Balancing R&D and production-level work
- Comfortable using AI-assisted coding tools (e.g. Claude Code, Cursor, GitHub Copilot) as part of your own workflow
- Judgment to govern and review AI-generated code responsibly
- Stay ahead of AI trends
- Continuously innovate by building and refining tools for internal and client use
- Actively apply cutting-edge research to improve processes and develop scalable, impactful solutions
- Proven track record of leading AI projects from concept to deployment
- Strong understanding of the AI development lifecycle
- Deep expertise in at least one part of the process
- Examples of deployed AI systems demonstrating agentic capabilities
- Experience with secure AI development practices
- Prompt injection awareness
- RBAC for agent systems
- AI governance within enterprise environments
- Strong verbal and written communication skills
- Fluent English, both written and spoken
Desired Qualifications
- Proven track record of leading AI projects from concept to deployment
- Strong understanding of the AI development lifecycle
- Deep expertise in at least one part of the process
- Examples of deployed AI systems demonstrating agentic capabilities
- Experience with secure AI development practices
- Prompt injection awareness
- RBAC for agent systems
- AI governance within enterprise environments
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.