What the job is
This job is for an AI Engineer who can work between research and building real-world solutions. You will help create and improve AI systems, from testing new ideas to making them work reliably and efficiently at a large scale. The role is fully remote and based in Johannesburg, Gauteng. You will work closely with product teams and software engineers to turn complex problems into useful AI features that people can actually use.

What you will do
You will design, build, and launch AI systems that use machine learning and large language models. This includes creating systems like retrieval-augmented generation (RAG) pipelines, custom fine-tuned models, and automated workflows. You will be responsible for the entire process of developing AI features, from collecting and preparing data to testing, deploying, and monitoring these systems once they are live.

You will also build and maintain the systems that support AI, such as data pipelines, embedding stores, and vector databases that help with search, retrieval, and personalizing user experiences. Another key part of the job is evaluating and using third-party AI tools like OpenAI or Anthropic, while also training and running open-source models on the company’s own servers. This includes preparing data, training models, optimizing them for efficiency, and making sure they run smoothly at scale.

You will need to keep up with the latest developments in AI and machine learning. This means testing new tools, models, and techniques, then sharing practical advice on what works best for the team. You will also work with AI coding assistants like Claude Code or GitHub Copilot to speed up development, but you will always review and verify the code they generate to make sure it is correct and safe.

What you need
You should have at least four years of experience building and deploying AI or machine learning systems in real-world settings. Strong skills in Python are a must, along with hands-on experience using frameworks like PyTorch or TensorFlow. You must also have practical experience working with large language models, including fine-tuning, RAG systems, or agent frameworks like LangChain or LlamaIndex.

Experience training and fine-tuning open-source models (such as Llama, Mistral, or Qwen) is essential. This includes full fine-tuning or efficient methods like LoRA/QLoRA, and deploying these models on your own infrastructure instead of just using cloud APIs. You should also be comfortable working with vector databases and embedding-based search tools like Pinecone, Weaviate, pgvector, or FAISS.

Knowledge of cloud platforms like AWS, GCP, or Azure is important, as well as experience with containerized deployments using Docker and Kubernetes, especially on GPU-powered servers. You should also be able to use AI coding tools effectively while still ensuring the quality and accuracy of the code they produce.

What you will get
This is a permanent, fully remote position with a salary of up to R800,000 per year, though this can be negotiated based on your experience. The role offers the chance to work on cutting-edge AI projects and contribute to innovative solutions in a dynamic environment. If you are interested, you can reach out to Nokuthula Gumbo at [email protected] or call 011 463 3633 to discuss the opportunity further.