Why I’m Building an AI Assistant for Everyone

Over the past six months working in technical support at a major telecommunications provider, I’ve had a front-row seat to how people experience technology when something goes wrong.

And one thing quickly became clear: the problem isn’t just technical , it’s human.

Every day, I speak with customers who are frustrated, confused, or simply tired of navigating systems that don’t seem designed for them. Some are highly technical and want fast, efficient solutions. Others — especially from older generations , just want clarity, patience, and reassurance.

Some want to talk. Some don’t.

That contrast is where this idea started.

The Inspiration: What Happens Between “Hello” and “Problem Solved”

In theory, customer support follows a simple flow:

1. Identify the issue
2. Ask the right questions
3. Apply the right fix

In reality, it’s much more complex.

People don’t always describe problems clearly. They skip steps. They misunderstand instructions. They get frustrated before the solution even begins.

After hundreds of interactions, patterns start to emerge. You begin to recognize recurring issues, predictable misunderstandings, and the small communication gaps that turn simple problems into long calls.

That’s when a question naturally came up:

What if part of this process could be handled automatically , without losing the human touch?

But Isn’t There Already an AI Assistant?

Yes, there is already an AI-based assistant in place.

However, being relatively new, it hasn’t yet earned the trust of all users. A significant number of customers, particularly those less familiar with technology , find it confusing, rigid, or impersonal.

The feedback is consistent:

• “It doesn’t understand me.”
• “It gives generic answers.”
• “I just want to talk to a human.”

This isn’t a failure of AI. It’s a mismatch between design and real user behavior.

My Goal: An AI Assistant That Adapts to the Human Not the Other Way Around

The project I’m developing is not about replacing human technicians.

It’s about building an intelligent support assistant that:

• Helps users who prefer not to call
• Reduces waiting times
• Solves simple, repetitive issues efficiently
• Adapts to the user’s level

What Will Make This Different?

1. Designed for All Levels of Experience

Most systems assume a certain level of digital literacy. Mine won’t.

Whether someone is:

• A tech-savvy user
• A casual smartphone user
• Or someone unfamiliar with basic settings

…the assistant will adjust how it communicates.

Simple language when needed. More direct instructions for advanced users.

2. Guided, Step-by-Step Troubleshooting

Instead of overwhelming users with information, the assistant will:

• Ask one question at a time
• Adapt based on responses
• Guide users through clear, structured steps

Think less “search engine,” more “interactive helper.”

3. Built from Real Support Experience

This isn’t based on theory or generic datasets.

It’s based on:

• Real conversations
• Real problems
• Real misunderstandings

The goal is to replicate the decision-making process of a technician , not just provide answers.

4. Respect for Frustration and Human Behavior

A key part often missing from AI systems: emotional awareness (at a basic level).

The assistant will be designed to:

• Recognize when users are stuck
• Simplify explanations
• Offer escalation to a human when needed

Because sometimes, the best solution is knowing when not to automate.

5. Multilingual and Accessible by Design

Starting with Greek and English, the system will aim to be:

• Easy to read
• Easy to follow
• Accessible across age groups

No unnecessary jargon. No assumptions.

The Vision

The goal is not to eliminate human interaction.

It’s to create a system where:

• Simple problems are solved instantly
• Complex problems reach technicians faster
• And every user ,regardless of age or technical knowledge, feels understood.

What Comes Next

This is just the beginning.

The next steps involve designing the system, building a working prototype, and testing it against real-world scenarios.

I’ll be sharing that journey , what works, what doesn’t, and what I learn along the way.

Because if there’s one thing working in technical support teaches you, it’s this:

Technology doesn’t need to become more complex. It needs to become more human.

This is just the start of the journey. In the next post, I’ll dive into how I’m designing and building this system step by step. Stay tuned.😊