Building and Deploying Intelligent Features with Generative AI

Turn generative AI into production-ready features

Using generative AI as a developer is one thing. Integrating it into software that other people can rely on is another.

On this advanced course, you’ll explore how large language models can become part of real applications through APIs, retrieval systems, intelligent features, and production-ready engineering practices.

You’ll start by connecting LLMs to applications using APIs, SDKs, and prompt orchestration.

From there, you’ll explore how capabilities such as assistants, semantic search, summarisation, and content generation can be shaped around genuine product and user needs.

Ground AI features in your own data

Explore retrieval-augmented generation (RAG), embeddings, and vector stores to connect language models with custom information.

You’ll examine how grounding can make AI features more relevant and useful while reducing reliance on a model’s training data alone.

Judge quality beyond whether it works

Assess intelligent features across reliability, output quality, cost, and latency. You’ll examine testing and monitoring approaches that help surface weaknesses before and after deployment and support better decisions about when an AI feature is ready for real use.

Operate generative AI in production

Explore versioning, observability, iteration, and ongoing maintenance for AI-powered features. You’ll also consider privacy, security, bias, safety, and governance across the product lifecycle.

By the end, you’ll be able to take generative AI beyond development assistance and apply it to the design, integration, deployment, and maintenance of intelligent features in real software applications.

This course is for software engineers, full stack developers, application developers, and technical leads ready to integrate generative AI into real products. Programming, prompt engineering, and AI-assisted development experience are recommended.

Learners need a computer, a stable internet connection, and access to the relevant development environments, APIs, SDKs, and AI tools demonstrated in the course. The course covers Gemini API, Groq API SDK, ChromaDB, Vertex AI, RAGAS, and container-based deployment. Specific operating system requirements, API access requirements, account requirements, and licensing details are not fully specified in the course material and should be confirmed before finalising the setup instructions.

This course is for software engineers, full stack developers, application developers, and technical leads ready to integrate generative AI into real products. Programming, prompt engineering, and AI-assisted development experience are recommended.

• Apply large language models in applications using APIs, SDKs, and tool-calling workflows.
• Create intelligent application features using multimodal capabilities, embeddings, vector stores, and retrieval-augmented generation.
• Evaluate intelligent features for quality, reliability, cost, and latency using appropriate evaluation methods.
• Modify AI-powered applications and their versions while maintaining reproducibility across prompts, models, outputs, and experiments.
• Improve monitoring, security, privacy, and responsible AI practices to operate intelligent features in production.