Opleiding Local Large Language Model Specialist (English online)

Module 1 — Foundations of Local & Hybrid AI
Learn:

The AI landscape in 2026
Local vs cloud vs hybrid AI
GDPR-first AI decision making
AI model evaluation
LM Studio
Docker fundamentals
AI infrastructure concepts
ROI-driven AI architecture decisions
You will understand:

Why fully local AI is not always optimal
Why fully cloud AI is often risky
How modern organisations combine both

Module 2 — RAG & AI Agent Orchestration
Learn:

Retrieval-Augmented Generation (RAG)
Vector databases with Qdrant
Embeddings and semantic search
Dify workflows
MCP (Model Context Protocol)
AI agents and tool calling
Multi-agent systems
Human-in-the-loop approvals
AI workflow debugging
Retrieval quality optimisation
You build:

A complete multi-agent onboarding platform
Connected AI systems with external tools
Production-style AI pipelines

Module 3 — Compliance, Fine-Tuning & AI Governance
Learn:

EU AI Act risk classification
AI System Cards
DPIA creation
Audit trails
AI governance workflows
Local model fine-tuning
LoRA and QLoRA concepts
Operational AI responsibility inside organisations

You finish with:

A defendable enterprise AI project
Full compliance documentation
A realistic deployment architecture
Technology Stack
You work hands-on with:

LM Studio
Dify
Qdrant
Docker Desktop
bge-m3 embedding models
Unsloth Studio (fine-tuning)
The course focuses on practical understanding instead of theory overload.

Online Learning Experience
This is a fully online, instructor-led training. The first and final lesson take place physically on campus in Antwerp. The campus is easily accessible by public transport and is located right next to Antwerpen-Berchem railway station.

The online edition was specifically redesigned for remote learning with:

Guided setup sessions
Interactive checkpoints
Live troubleshooting
Prebuilt training environments
Recorded sessions
Structured peer feedback
Incremental project milestones
Students never start from scratch during lessons.
Every session builds on a prepared environment to maximise learning efficiency.

Final Project & Defence
At the end of the course, you present your own AI implementation project.

You apply the same architecture principles to:

Your company
Your department
Your own operational use case
The final defence includes:

Technical architecture
Compliance reasoning
AI governance decisions
Deployment strategy
Human oversight design

Information and registrations are available through the colleagues at Syntra AB: Local Large Language Model Specialist (e-learning and in English).

This course is ideal for:

IT professionals
AI project leads
Technical consultants
Innovation managers
Data & automation professionals
Internal AI champions
Solution architects
SME technology decision makers
Especially valuable for organisations that:

Cannot send sensitive data to public AI platforms
Need GDPR-compliant AI workflows
Want to deploy private AI internally
Need practical AI governance knowledge
Want to understand local AI infrastructure
Prerequisites
You do not need programming experience.

You should:

Be comfortable with IT concepts
Be willing to work hands-on
Have a modern laptop capable of running Docker and local models
A complete setup session is included before the first lesson.

During your training, you are free to use AI tools as support for learning and practice. We believe it is important that you learn how to work with AI in the same way it is used in the professional field, especially within IT-related programmes where it has become an essential part of the workflow.

You are expected to use your own AI subscription. Any additional costs, such as AI credits required by certain tools, are at your own expense. Please refer to our AI policy for full details.

The course itself is taught in English. However, administrative communication such as emails, official letters and certificates will be provided in Dutch.

Our online classes are delivered as live sessions only — no recordings are made, including in cases of absence or illness. This is a deliberate choice, as active participation and interaction are essential for properly understanding and applying the course material.

Attendance during lessons is therefore important and has a direct impact on both your learning progress and your final project.

After completing this course, you will be able to design and deploy secure AI workflows using local LLMs, RAG pipelines, AI agents and hybrid AI architectures in a realistic business environment. You will understand how to connect AI systems to company data and tools, evaluate when data can remain local or move to the cloud, and build AI solutions that align with GDPR and the EU AI Act — enabling you to translate AI from experimentation into practical organisational value.