How GenAI works — Through the Eyes of an Engineer |Practical Pocket Guide
Last Updated on July 27, 2026 by Editorial Team
Author(s): Stylianos Chiotis
Originally published on Towards AI.
If you’re not a Medium member and would like to enjoy this article without restrictions, feel free to use the friend link here. Happy reading!
This is how an LLM actually works — end to end, no any mysticism!
The author frames LLMs as an engineered “data pipeline” rather than magic: user input becomes a prompt, is tokenized into numbers, mapped to embeddings (vector representations), and processed by a transformer “black box” that reasons via probability while accounting for word order through positional encoding. They explain why tokens affect cost (measured in input/output tokens) and how tokenization is deterministic and tokenizer-dependent, so the same text can have different token counts under different tokenizers. The discussion then zooms out to the model’s generation loop (predicting next tokens until an EOS or a max token limit), detokenization back to human language, and the role of learned parameters and transformer layers in forming capabilities. The article closes by connecting these mechanics to how people should interpret model outputs (as probabilistic guesses, not absolute truth), encouraging experimentation with open-source models, and emphasizing learning through questions and debate.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Recent Posts
