Bulien Logo

Making Sense of AI

There’s a lot of noise out there, so here’s a simple breakdown of some of the terms you might’ve heard:

LLM (Large Language Model)

  • Think of this as a very smart text engine that’s been trained on loads of information. It can answer questions, summarise documents, write content, and more.
  • It powers services like ChatGPT and other smart assistants.

MCP (Model Context Protocol)

  • A standard way to connect AI models to your data and tools, without needing endless custom code.
  • Helps you build AI solutions that are more accurate and faster to implement.

Agent (Agentic AI)

  • This is when an AI tool doesn’t just answer questions – it takes action. For example, checking your calendar, sending an email, or starting a task.
  • It’s like having a digital assistant.

Tools We Use

We use a mix of tools depending on the job:

  • OpenAI (GPT-4o) – great for language-based tasks like summarising text or generating content.
  • Google Gemini – a strong option for working with text, images and audio together.
  • Meta Llama 3 – useful for building AI tools in private environments.
  • Anthropic Claude – for long documents and compliance-conscious projects.
  • DeepMind Veo – very powerful tool for video and creative tasks.
  • n8n – connect AI into your existing tools, like CRMs or emails.
  • Databricks & Alteryx – amazing platforms we use for data cleaning, reporting, and pipeline automation.

Where We’ve Used AI

We’ve built an agent to monitor real-time sensor data and trigger proactive maintenance alerts before failures occurred.

We’ve used multiple LLMs to drive smarter product recommendations and improve personalisation across marketing channels.

We’ve developed a natural-language search assistant that surfaces guidance from thousands of internal documents instantly.

We’ve created a compliance workflow that checks financial transactions against sanction lists and auto-generates reports.

We’ve automated internal ticket routing and triage by using an agent to classify and assign incoming support requests.

We’ve integrated AI into regular reporting processes to summarise changes, flag anomalies, and suggest next actions.

We’ve…. done a lot more than that – ask us, we’d love to talk to you about it.

How We Help

Practical AI Consulting

We’ll sit down with you, understand what you’re trying to achieve, and help you figure out whether AI makes sense – and if so, where to start.

 

Upskilling and Support

Whether it’s a quick proof of concept or a full training session with your team, we make sure that AI isn’t something only the IT team understands. We’ll get your people confident and capable.

 

Ongoing Partnership

We don’t just disappear after the first delivery. We stick around as long as you need us – helping you grow, improve and get more value out of what you’ve built.

Let’s chat about how AI could work for you.