Applied Generative AI and Production Systems

Gain advanced generative AI skills

Taking generative AI from experimentation to production requires more than effective prompting. On this three-week course, you’ll learn how to specialise models, build multimodal AI applications, and apply the engineering practices needed to operate generative AI systems reliably.

You’ll explore when fine-tuning is more appropriate than prompting or retrieval, then use parameter-efficient techniques including LoRA, QLoRA, and adapters through the Hugging Face ecosystem. You’ll prepare instruction-tuning datasets and evaluate model performance before and after fine-tuning.

Build and deploy multimodal AI applications

Extend your generative AI capabilities beyond text. You’ll work with vision-language models for image captioning and visual question answering, as well as speech-to-text and text-to-speech technologies.

You’ll then learn how to deploy generative AI applications using FastAPI, containerisation, and cloud platforms. By combining these technologies, you’ll develop the skills to turn AI models into usable applications that can operate beyond a development environment.

Learn to manage AI systems in production

Discover the practices that help organisations operate LLM applications effectively throughout their lifecycle. You’ll explore LLMOps, including prompt and output versioning, experiment tracking, tracing, observability, and model drift monitoring.

You’ll also learn how to evaluate and improve AI systems. The course culminates in a project where you’ll build, deploy, evaluate, and monitor a production-ready generative AI application.

By the end, you’ll have practical experience that will help you progress towards roles such as Generative AI Engineer, LLM Engineer, and Applied AI Engineer.

This course focuses on specialising models, building multimodal systems, and running GenAI reliably in production. It is for learners who can build LLM applications and RAG systems and who now want to adapt models to specific tasks and operate them dependably.

It is ideal for aspiring GenAI engineers and LLM engineers, and developers preparing for roles where AI systems must be fine-tuned, deployed, monitored, and maintained. Learners should be confident with LLM APIs, frameworks, and RAG before starting. By the end, they will have built, deployed, and monitored an end-to-end GenAI application and will be ready for generative AI engineer roles.

This course focuses on specialising models, building multimodal systems, and running GenAI reliably in production. It is for learners who can build LLM applications and RAG systems and who now want to adapt models to specific tasks and operate them dependably.

It is ideal for aspiring GenAI engineers and LLM engineers, and developers preparing for roles where AI systems must be fine-tuned, deployed, monitored, and maintained. Learners should be confident with LLM APIs, frameworks, and RAG before starting. By the end, they will have built, deployed, and monitored an end-to-end GenAI application and will be ready for generative AI engineer roles.

• Describe the roles of fine-tuning, prompting, and RAG in generative AI model specialisation.
• Experiment LoRA, QLoRA, and adapter-based fine-tuning using the Hugging Face ecosystem.
• Assess GenAI systems using model evaluation, guardrails, red-teaming, tracing, and drift monitoring.
• Design an end-to-end production GenAI solution incorporating specialisation, deployment, safety, observability, and LLMOps practices.
• Compare multimodal and deployment approaches involving vision, speech, FastAPI, Docker, and cloud platforms.