Skip to content
← All courses
Teams

Local AI training: your data stays with you

Install and use AI on your workstation, without sending data to the cloud. Ollama, AnythingLLM, RAG on your documents. Digital sovereignty starts here.

1 day (6 h 45 of training) From €1,800 excl. VAT per day OPCO or FIF-PL funding possible

Trainee satisfaction: 4/4 (post-session evaluations, Qualiopi records)

Target audienceProfessionals handling sensitive data: regulated professions, finance, legal, HR, healthcare

What you'll be able to do

Understand how a local LLM works and its privacy advantages

Use Ollama to interact with different AI models locally

Choose the right model for each type of task

Use an AI agent to interact with local documents via RAG

Apply prompting best practices for relevant results

Programme

  1. Session 1

    Why local AI, and how to set it up

    • Cloud or local: what goes online and what stays with you
    • Installing Ollama and downloading models
    • Installed models: Mistral (French), Qwen 3.5 (reasoning)
    • First test: asking a question to a local model
  2. Session 2

    AnythingLLM: AI with a real interface

    • Installing and configuring AnythingLLM
    • Creating your first workspace
    • Asking business questions in the interface
    • Comparing models: Mistral vs Qwen
  3. Session 3

    Writing effective prompts

    • The winning structure: context → role → task → format → constraints
    • Before/after examples on your use cases
    • Configuring a system prompt in AnythingLLM
    • Writing 5 optimised prompts for your recurring tasks
  4. Session 4

    Working on your documents with RAG

    • RAG: AI reads your documents and answers with sources
    • Drag and drop a full folder into a workspace
    • Querying documents: extraction, summary, cross-referencing
    • Producing a summary note from multiple documents
  5. Session 5

    Organising daily use

    • Creating a workspace for each client folder
    • Decision tree: local or cloud by data sensitivity
    • Daily launch procedure
    • Demo of advanced tools: Claude Code + Ollama, AnythingLLM

Practical information

Duration

1 day (6 h 45 of training)

Target audience

Professionals handling sensitive data: regulated professions, finance, legal, HR, healthcare

Prerequisites

Regular use of a computer (Mac or PC). No programming skills required.

Group size

1 to 4 people

Pricing

From €1,800 excl. VAT per day

OPCO funding possible.

Teaching approach

100% hands-on on the trainee's workstation. The trainer installs and configures, the trainee learns to use. Exercises on real or fictional documents adapted to the role.

Assessment

Entry assessment, one assessed exercise per session. At the end, a before-and-after self-assessment of skills, then the trainer validates each learning objective (acquired, in progress, not acquired), recorded in the end-of-training certificate.

Trainer

An AI consultant from Colombani.ai.

Booking lead time

At least 2 weeks between enrolment and the start of training.

Funding

Qualiopi-certified provider (certificate QUA009472), registered under training provider number 11757549975 (this registration does not constitute State approval): OPCO or FIF-PL support can be requested. The decision rests with the funding body, based on your sector, its criteria and its budgets. Colombani.ai prepares the application with you.

From €1,800 excl. VAT per day for an in-company group; final price depends on group size and length, on quote. OPCO or FIF-PL funding can be requested depending on your status; the amount left to pay depends on the sum granted by the funding body.

Discuss this course (30 min) OPCO funding is discussed during the call.

Receive the programme and a quote

Tailored programme and quote within two business days.

Accessibility

This course is accessible to people with disabilities. Contact the disability officer in advance to discuss possible adjustments.

Ulysse Trin, referent-handicap@colombani.ai

Post-training support

✓ Installed and configured environment: Ollama + models + AnythingLLM
✓ Personalized getting started guide (PDF)
✓ AnythingLLM workspace configured with adapted system prompt
✓ Training completion certificate

Frequently asked questions

Do I need a powerful computer to run local AI? +

A recent computer with 8 GB of RAM is enough for lightweight models (Mistral 7B, Qwen 3.5). For larger models, 16 GB is recommended. We check compatibility beforehand.

Does my data really stay on my machine? +

Yes, 100%. Ollama and AnythingLLM run entirely locally. No data is sent over the internet. That's the whole point of local AI.

Is local AI as powerful as ChatGPT? +

For most business tasks (summarisation, extraction, writing), local models are more than adequate. The training teaches you to choose the right model for each task.

Receive the programme and a quote

Tailored programme and quote within two business days.

Got a project in mind?

Describe your situation. The first 30-minute call is free. Get a frank response within one business day.