Deploying LLMs On-Premise: Hardware, Model Choice and Cost
You can use large language models without your data leaving the building. Model choice, GPU sizing, serving software, security and the cost case for on-premise LLM deployment.
Read more →
Technical Teams
Building LLM Applications
RAG architectures, agents and MCP, output evaluation, security and on-premise LLM deployment.
- 3 days
- Software, data and infrastructure teams
- On-site or online
- Certificate of attendance
What you will learn
- Design the architecture of LLM-based applications and choose the right model
- Build a RAG system over corporate documents
- Build tool-using agents and connect them to enterprise systems with MCP
- Evaluate output, defend against security threats and plan on-premise deployment
Course content
- How LLMs work; choosing closed vs. open-weight models; API usage
- Prompt and context engineering, structured output
- RAG: document processing, vector databases, hybrid search and citations
- Agents, tool use and the Model Context Protocol (MCP)
- Evaluation (LLM-as-a-judge), security (prompt injection, guardrails) and on-premise serving with vLLM
Who should attend
- Software developers
- Data scientists and ML engineers
- Teams building GenAI products
Prerequisites
Software development experience with Python.
Course content is tailored to your organisation's needs. All programmes can be delivered on-site at your premises or online. Participants who complete a programme receive a certificate of attendance.
Frequently Asked Questions
Which models are used?
Both closed and open-weight models. The choice depends on whether you prefer cloud or on-premise.
Does it cover on-premise LLM deployment?
Yes. Serving models with vLLM and the basics of on-premise deployment are part of the programme.
What will participants have built by the end?
Working examples such as a RAG application over company documents and a simple tool-using agent.
Is security covered?
Yes. Prompt injection, guardrails, access control and output evaluation are core topics.
Free assessment · 2 min
Start now
Training Finder: Which AI Training Is Right for You?
Answer a few short questions and see the AI training programmes that suit your team as an ordered path, with the reason for each.
Related articles
What Is RAG? Building an AI Assistant on Your Company Documents
RAG lets language models answer based on your organisation's own documents. How it works, how it differs from fine-tuning, what to watch when building it and how to measure quality.
Read more →Which AI Training Do Your Employees Need? A Role-Based Guide
Executives, HR specialists, analysts and developers all need different AI training. A practical guide to matching each role with the right programme.
Read more →Related courses
Data Science with Python
Data preparation, exploratory analysis and visualisation: a solid foundation in data science.
Learn more →Applied Machine Learning
Classification, regression, model evaluation and feature engineering on real-world datasets.
Learn more →MLOps: Taking Models to Production
CI/CD, model versioning, monitoring and data drift, plus LLMOps for LLM applications: keeping models reliable in production.
Learn more →DevOps & AWS Fundamentals
Containers, automation and the fundamentals of AWS cloud architecture.
Learn more →Get a tailored training proposal
Tell us which programmes you are interested in, the number of participants and the preferred format, and we will prepare a tailored proposal.