Skip to content
Neural Soft
Contact us Español (View this page in Spanish)

Building with AI

LLM application development course

From the first API call to a generative AI application ready for production.

Request information

Integrating a language model into an application is easy; doing it reliably, securely and at a controlled cost is something else. This course is for developers who want to move from experiments to features that can go into production.

We cover everything you need: the API in depth, streaming, validated structured outputs, tool use and MCP, basic RAG, architecture, caching, automated tests, security and observability. With code from the first module and a complete project at the end.

What you will learn

  • Integrate language models into real applications through their APIs
  • Build features with structured outputs, tools and streaming
  • Design the architecture of an LLM application
  • Control cost, latency and errors
  • Deploy securely and with observability

Syllabus

9 modules, from the fundamentals to the final project.

  1. Fundamentals

    • How an LLM works, for programmers
    • Models, providers and selection criteria
    • Pricing, limits and quotas

  2. The API in depth

    • Messages, roles and system instructions
    • Streaming responses
    • Image and document inputs

  3. Structured outputs

    • JSON schemas and validation (Pydantic, Zod)
    • Data extraction and classification

  4. Tools

    • Function calling step by step
    • Connecting to APIs and databases
    • Model Context Protocol (MCP)

  5. Your own knowledge

    • Basic RAG with embeddings
    • Long context versus retrieval

  6. Architecture

    • Application patterns with LLMs
    • Queues, retries and timeouts
    • Prompt and response caching

  7. Quality

    • Automated tests and evaluation sets
    • Comparing models and versions

  8. Security

    • Instruction injection and data leakage
    • Risks in the OWASP Top 10 for LLM applications
    • Managing keys and personal data

  9. Production

    • Observability and tracing
    • Cost control per user and feature
    • Changing models without surprises

Tools you will practise with

  • Python or TypeScript
  • OpenAI, Anthropic, Google or Azure OpenAI API
  • Pydantic or Zod
  • FastAPI or Node.js
  • Langfuse

Hands-on project

You develop a small complete application, for example a service that receives documents, extracts their data with structured output, calls an API through tools and returns the result, with tests, tracing and cost control.

Requirements

  • Experience programming in Python or TypeScript.
  • Knowledge of REST APIs.

Other courses

Where it is applied

  • AI integration in apps and systems

    We add AI to the tools your business already uses to automate tasks, search information and connect data without having to change everything.

  • Custom AI development

    We design and build artificial intelligence solutions tailored to your processes, data and tools when an off-the-shelf solution does not fit or falls short.

Want this course for your team?

We adapt the content, length and format to your company.

Request information