Generative AI for Developers

Large language models have become a normal part of the developer toolkit, both as a feature you build into products and as an assistant that helps you write code. This course focuses on the developer side: how to call model APIs from Python or TypeScript, design prompts, get reliable structured output, connect models to your data and tools, and ship features that are tested and safe.

You start with how LLMs work at a practical level (tokens, context windows, temperature, cost and latency), then build progressively: chat and streaming interfaces, structured JSON output, function or tool calling, retrieval-augmented generation with embeddings and a vector store, and simple agents that use tools. Model Context Protocol (MCP) is introduced as a way to connect models to external tools and data. You also learn evaluation, prompt-injection defences, privacy considerations and cost control. A separate module covers using AI coding assistants well: writing good prompts for code, reviewing output, and keeping code quality high.

There is no external vendor exam for this course. Model providers and their features change quickly, so the course teaches concepts that stay stable and uses current provider APIs for hands-on work. You receive an Assure Success course-completion certificate on completion.