Generative AI for Developers

Overview

Generative AI for Developers focuses on engineering applications on large language models. You will work through a model comparison harness, a resilient chat client, a…

What you will learn

  • Explain tokens, context windows, and inference parameters and select a model by task, cost, and latency (Module 1).
  • Integrate an LLM API with streaming, retries, rate-limit handling, and logging (Module 2).
  • Manage prompts as versioned code and return validated structured output (Module 3).
  • Generate embeddings and query a vector store with metadata filters (Module 4).

Prerequisites

  • Required: Python fundamentals and basic REST API knowledge. A laptop able to run Python, Docker, and an IDE. API access to at least one LLM provider [confirm whether credits are provided].
  • Helpful: Git, basic SQL, and prior use of any AI assistant. If you are new to Python, complete a Python foundation course first.

Curriculum

  1. Module 1: Generative AI Foundations for Engineers

    • Topics: Transformers and attention at a high level; tokenization and embeddings; context windows; proprietary vs open models; choosing by task, cost, and latency; multimodal models.
    • Deliverable: Comparison report. Completion check: Module 1 assessment.

    Hands-on:

    • Practice: Model Comparison Harness. Three models, same prompts, measured quality, latency, and cost.

  2. Module 2: Working with LLM APIs

    • Topics: Messages, roles, and system prompts; temperature, top-p, max tokens; streaming; retries, timeouts, rate limits; caching and idempotency; logging and observability.
    • Deliverable: Chat client code. Completion check: Module 2 assessment.

    Hands-on:

    • Practice: Resilient Chat Client. Streaming client with retries and token tracking.

  3. Module 3: Prompt Engineering in Code

    • Topics: Parameterized templates; few-shot in code; prompt files in version control; JSON mode and schemas; validation with Pydantic; handling malformed output.
    • Deliverable: Classifier service. Completion check: Module 3 assessment.

    Hands-on:

    • Practice: Ticket Classifier. Validated JSON with confidence scores.

  4. Module 4: Embeddings and Vector Search

    • Topics: Semantic similarity; chunking strategies; embedding model choice; indexing and metadata filtering; hybrid search; scaling considerations.
    • Deliverable: Search API. Completion check: Module 4 assessment.

    Hands-on:

    • Practice: Semantic Document Search. Index a document set and expose a search API.

  5. Module 5: Retrieval-Augmented Generation

    • Topics: Ingest, retrieve, generate; citations and grounding; context assembly; re-ranking; query rewriting; evaluating retrieval.
    • Deliverable: RAG assistant. Completion check: Module 5 assessment.

    Hands-on:

    • Practice: Policy Q&A Bot. RAG assistant over policy documents with citations.

  6. Module 6: Function Calling and Agents

    • Topics: Function schemas; tool selection and execution loop; tool error handling; ReAct and planning loops; memory and state; guardrails and stopping conditions.
    • Deliverable: Agent code. Completion check: Module 6 assessment.

    Hands-on:

    • Practice: Data Assistant Agent. Agent calling database and weather tools.

  7. Module 7: Evaluation, Safety, and Cost

    • Topics: Golden datasets; LLM-as-judge; regression testing; prompt injection defenses; PII handling; token budgeting and caching.
    • Deliverable: Evaluation suite. Completion check: Module 7 assessment.

    Hands-on:

    • Practice: Eval Pipeline. Score the RAG bot and flag regressions.

  8. Module 8: Deploying Generative AI Applications

    • Topics: API layer and UI; async processing; secrets and configuration; monitoring and tracing; feedback loops; model updates and fallbacks.
    • Deliverable: Deployed application. Completion check: Module 8 assessment.
    • Capstone (independent work, [CAPSTONE_HOURS] hours): An end-to-end generative AI application combining RAG, tool calling, structured outputs, an evaluation suite, and a cost and safety report.

    Hands-on:

    • Practice: Ship It. Containerize and deploy the RAG assistant with a web UI and monitoring.

What's included

  • Live cohort + recordings: Attend every session live and revisit the recordings for one year.
  • Real lab environment: Practise on a hands-on environment that mirrors real projects.
  • Live doubt-clearing support: Ask questions and get answers in real time during every session.
  • Assessments and lab activities: Apply what you learn through graded assessments and guided labs.
  • Learning assets: Keep the slides, notes, and reference material for good.
  • Role-readiness guidance: Mentor guidance to get you ready for the role you are targeting.
  • Verifiable certificate: Earn a shareable certificate on completing the hands-on capstone.
  • Global peer community: Learn with a global cohort and stay connected after the program.

In short

Generative AI for Developers is a 24-hour instructor-led course covering Module 1: Generative AI Foundations for Engineers, Module 2: Working with LLM APIs, Module 3: Prompt Engineering in Code. It includes guided practical labs.

Key facts

  • Course: Generative AI for Developers
  • Duration: 24 hours
  • Modules: 8
  • Format: instructor-led live training

Frequently asked questions

How much Python do I need?

Comfortable with functions, classes, packages, and HTTP requests. You do not need data science or ML experience.

Which model providers are used?

OpenAI and Anthropic APIs for hosted models and Hugging Face for open models. Patterns transfer to other providers.

Will API costs be covered?

[Confirm]. Lab usage is kept small; typical spend for the full course is modest but depends on provider pricing.

Does this cover LangChain?

Introductory use only. For deep LangChain and LangGraph work, take RAG, LangChain & LangGraph after this course.

Do I deploy to the cloud?

You containerize and deploy locally or to a free-tier host of your choice. Cloud account costs, if any, are yours.

Is this the same as LLMOps & Production GenAI?

No. This course builds the application. LLMOps covers operating many such applications: registries, CI evaluation gates, serving, and cost optimization at scale.

What do I keep?

A full repository with RAG, agent, evaluation suite, and deployment configuration.

Is a corporate version available?

Yes, including a reference build on your approved stack.

Topics covered

Generative AI for Developers, Artificial intelligence, Corporate training