AI Engineering

The final level develops the ability to design and build modern AI applications using LLMs, embeddings, RAG, agents, evaluation, and deployment concepts.

Training Modules

Modern AI Models

• Practical neural network foundations.
• Embeddings and semantic representations.
• Transformers, LLMs, tokens, and context.

Building LLM Applications

• Working with LLM APIs.
• Structured outputs and tool use.
• Context management and multi-step workflows.

RAG and Knowledge Retrieval

• Chunking and embeddings.
• Vector databases and semantic search.
• Designing RAG pipelines connected to knowledge sources.

Agents and Quality

• AI agents and tool use.
• Multi-step workflows and orchestration.
• Evaluation, guardrails, safety, and reliability.
• Deployment and monitoring fundamentals.

Practical Work

Build functional AI components and connect models with data, tools, and interfaces while evaluating and improving results.

Capstone Project

Design an integrated AI application such as a knowledge assistant or document Q&A system using LLMs, RAG, evaluation, documentation, and a final presentation.

Expected Outcomes

• Build applications using LLM APIs.
• Design RAG systems connected to knowledge sources.
• Use embeddings and vector search effectively.
• Evaluate AI system quality and safety.
• Deliver an integrated AI project.

Prerequisites

• Successful completion of Levels 1 and 2 is mandatory.
• Direct entry to Level 3 is not permitted.
• Completion requires passing the assessments and capstone project.