Generative AI for developers: tools and applications

Module 1: Understanding generative AI fundamentals

• Artificial intelligence (AI) basics: definitions and main categories (symbolic AI, supervised learning, unsupervised learning and deep learning).
• Large language model (LLM) architecture and how Transformers work.
• Differences between generative AI and other traditional AI approaches.
• Major generative AI models (GPT, Claude, LLAMA, Mistral and DALL-E).
• Development use cases (code generation, documentation, testing, conversational interfaces, semantic analysis, prototyping and asset generation).

Practical exercises:

• Test and explore several models through their interfaces (ChatGPT, Claude, Hugging Face Spaces, DALL-E, Playground, etc.).

Module 2: Mastering prompt engineering

• The anatomy of a prompt (instruction, context and expected format).
• Advanced prompt engineering techniques (zero-shot, few-shot, chain-of-thought and system messages).
• Specific prompts for ideation, specification, design, coding, testing and deployment.
• Using complementary tools such as LangChain and PromptLayer.

Practical exercises:

• Create a library of technical prompts (code generation, summarisation, testing, JSON–HTML conversion, etc.).

Module 3: Integrating generative AI through APIs

• Using generative AI APIs (OpenAI, Hugging Face and Replicate for text-to-text, image and code).
• API request parameters (temperature, max_tokens, top_p, frequency, etc.).
• Managing API keys, quotas, logs, rate limits and costs.

Practical exercises:

• Create a developer assistant (code or documentation generator) through an API with a simple interface (Python Flask or React front end).

Module 4: Exploring practical use cases and automation

• Automatically generating technical documentation from source code.
• Writing tickets or changelogs from commits.
• Generating usage examples or unit tests.
• Error detection and assisted code review.
• Generating user interfaces (UI) from descriptions.

Mini-project — AI challenge:

• Take on the challenge of creating a small generative AI tool, choosing from examples to guide you:
  • Develop a chatbot for technical documentation;
  • design a customised code snippet generator;
  • develop an intelligent code review tool;
  • create a wireframe or interface generator from text descriptions.

Module 5: Managing generative AI security and regulation

• Generative AI limitations, risks and ethical governance (hallucinations, bias, GDPR, copyright and security).
• Methods for governing generative AI use in professional environments.