Cisco Networking Academy: Exploring Modern AI

Over the past few weeks, I’ve been working through the Introduction to Modern AI course from Cisco Networking Academy and have just completed it with a score of 90%!

This course felt particularly relevant given how quickly AI is evolving and becoming embedded in everyday tools, workflows and decision-making processes.

Building a Strong Foundation

What I appreciated most about this course is how it broke down complex ideas into clear, structured concepts.

It started with the fundamentals:

• What artificial intelligence actually is (beyond the buzzwords)
• How machine learning enables systems to learn from data
• The role of models, parameters and training vs inference

Revisiting ideas like the Turing Test and early symbolic AI gave useful context for how far the field has come and why modern approaches, particularly machine learning, have become so dominant.

From Theory to Real-World Applications

As the course progressed, it explored how these concepts apply in practice.

Topics like computer vision stood out, especially understanding:

• Object detection vs image classification
• How image segmentation enables effects like portrait mode
• The role of embeddings in similarity search

It’s one thing to use these features every day in apps, but another to understand what’s actually happening behind the scenes.

Understanding How Chatbots Actually Work

One of the most interesting sections focused on chatbots and large language models (LLMs).

A key takeaway for me was this:

A chatbot takes your input, the prompt, and it does not output its entire completion all at once. It actually outputs a word or part of a word at a time.

That simple idea explains a lot:

• Why prompting matters: responses are built step by step, so the input shapes what comes next.
• Why reasoning-first prompts often produce better answers: asking for reasoning before an answer helps guide later outputs in the sequence more effectively.
• Why hallucinations can happen: each word is generated without a full, verified ‘overview’ of the final answer.

This also reinforces the importance of being intentional when using AI tools by structuring prompts clearly, breaking tasks into steps and always validating outputs.

Practical Skills To Take Away

Beyond the theory, there were a lot of insights that can be applied in practice:

• Writing more effective, structured prompts
• Breaking complex problems into smaller steps
• Using multiple models or tools depending on the task
• Understanding the limitations, not just the capabilities, of AI systems

The idea of ‘thinking like a detective’ when working with AI really stuck with me: testing outputs, questioning results and refining inputs.

A Broader Perspective on AI

The course also touched on bigger-picture ideas:

• The difference between generative (GenAI) and non-generative AI
• The role of foundation models and transformers
• Concepts like Retrieval Augmented Generation (RAG)
• Trade-offs between model size, performance and sustainability

It highlighted that AI is not a single tool but an entire ecosystem of various models, approaches and use cases.

Final Thoughts

Completing this course has given me a much clearer understanding of how modern AI systems work, not just the theory but also how it works in practice.

With AI continuing to shape the future of technology, products and decision-making, building foundational knowledge within the field feels like an important step to not get left behind in our dynamic, ever-changing world.

Much like my MSc journey, my exploration of emerging technologies and AI has reinforced something I have observed throughout my professional career and academic studies: there is always more to learn and that constant evolution is what makes the modern world so interesting.