Jobs at Temus (Now Hiring) — 4 open

Your Role

• Harness your expertise in AI engineering, delivery, and governance to solve complex problems and deliver transformative change for clients and citizens.
• Make an outstanding contribution to a world-class AI and data team.
• Develop and maintain relationships with a broad range of clients, colleagues, and partners across a variety of contexts and formats.
• Invest continuously in building and extending your knowledge and skills.

Your Background

• At least 3 years of relevant working experience in fields related to artificial intelligence, machine learning, data science, or software engineering.
• Completed at least one hands-on AI development and deployment end-to-end.
• Exposure to agentic AI concepts such as multi-agent systems, tool-use, RAG pipelines, or autonomous task execution is highly desirable.
• At least 3 years of implementation experience with serverless computing, CI/CD, containerisation, Infrastructure as Code, code version control and automated testing.
• Exposure to one or more cloud services: AWS, Azure or GCP.
• Candidates should demonstrate strength in either AI Engineering or ML Engineering (see below). Strength in both is a plus.

AI Engineering (LLM & Agentic Systems)

• Practical hands-on experience with LLMs and agentic tooling: LangChain, LangGraph, AutoGen, CrewAI, OpenAI API, Anthropic API, AWS Bedrock, Google Vertex AI, Azure ML, Hugging Face Transformers, MLFlow, Dataiku, MS Fairlearn, Google PAIR.
• Experience with prompt engineering, fine-tuning, and responsible AI evaluation.

ML Engineering / MLOps

• Practical hands-on experience with ML development and deployment tooling: Jupyter, PyTorch, TensorFlow, Scikit-learn, AWS SageMaker, MLFlow, Docker, Kubernetes, Terraform, Ansible, Datadog.
• Broad knowledge of NLP, computer vision, recommendation systems, reinforcement learning and time series forecasting.
• Exposure to model experiment tracking and training workflows: hyperparameter tuning, experiment logging (e.g. MLFlow, Weights & Biases), and reproducible training runs.
• Exposure to the MLOps lifecycle: data versioning, model training pipelines, experiment tracking, model registry, deployment, monitoring, and retraining triggers — using tools such as MLFlow, Weights & Biases, Kubeflow, AWS SageMaker Pipelines, Azure ML Pipelines, or Google Vertex AI Pipelines.