| Agentic Loops | Build the loop that runs an agent (execute tool calls, feed results back, stop safely) with the limits and instrumentation that keep it reliable in production. | 2026-10-21 |
| AI Agents Explained | Understand an agent as a model in a loop calling tools, so you can tell when you need one and which failure modes to design for. | 2026-10-21 |
| Context Windows and Management | Budget what goes into each model call (system prompt, history, retrieved docs, tools, output) so answers stay accurate and cost stays predictable. | 2026-10-21 |
| Diffusion Models | Diffusion models make images (and video and audio) by starting from pure random noise and removing it a little at a time, guided by your text prompt; this page explains the idea, why it works, how text steers it, and what the main variants are, with a small runnable example of the noising process. | 2026-10-21 |
| Embeddings and Vector Search | Learn how text becomes vectors and how nearest-neighbor search over them powers semantic search and RAG, so you can choose models, chunking and indexes deliberately. | 2026-10-21 |
| Evals for LLMs | Build a test suite for your LLM app (representative inputs, a definition of good, automated scoring) so prompt and model changes are measured instead of guessed. | 2026-10-21 |
| Fine-Tuning vs RAG | Decide between fine-tuning (change how the model behaves) and RAG (change what it knows), and default to RAG unless you can say why not. | 2026-10-21 |
| GPUs for AI | Understand why AI runs on GPUs and how memory, not raw compute, usually decides which GPU (if any) you need. | 2026-10-21 |
| Guardrails and Safety | Layer input checks, output checks, policy and runtime limits around an LLM app, knowing which layers actually stop harm and which only look like they do. | 2026-10-21 |
| Hallucinations | Language models sometimes state false things fluently and confidently; this page explains why that happens (it is a predictable result of how they are trained and graded, not a random glitch), which tasks are most at risk, and the concrete habits and system designs that reduce the damage. | 2026-10-21 |
| Inference Servers | Pick and tune an inference server (vLLM, TGI, SGLang, llama.cpp and others) when you self-host open-weights models, and know when a hosted API is cheaper. | 2026-10-21 |
| LLM Basics | Learn what a large language model actually does (predict the next token) so every other AI concept on this site has something to hang on. | 2026-10-21 |
| MCP Explained | Learn what the Model Context Protocol standardizes (tools, resources, prompts) so you can write one server and plug it into any MCP client. | 2026-10-21 |
| Multimodal Models | Learn how models take images, audio, video and PDFs alongside text, and when a multimodal call beats a dedicated OCR or speech pipeline. | 2026-10-21 |
| Neural Networks in Plain English | A neural network is a big stack of simple adjustable calculations that learns by nudging millions or billions of numbers to reduce its mistakes; this page explains neurons, layers, weights, training and why scale matters, with no calculus and one small network you can run. | 2026-10-21 |
| Prompt Caching | Turn on prompt caching for any stable prompt prefix to cut input cost and latency, and structure prompts so the cache actually hits. | 2026-10-21 |
| Prompt Engineering | Write prompts that work reliably: clear instructions, structure, examples and constraints, tested against a small eval set rather than by feel. | 2026-10-21 |
| Quantization and Distillation | Shrink a model by storing weights at lower precision (quantization) or training a smaller student (distillation), and measure the quality you lose on your own task. | 2026-10-21 |
| RAG Explained | Learn how retrieval-augmented generation feeds your own data into a prompt at query time, and the failure modes to test for before you ship it. | 2026-10-21 |
| Structured Outputs | Force model output to match a JSON schema when it feeds code, and know when schema-constrained output hurts quality instead of helping. | 2026-10-21 |
| Tool Use and Function Calling | Learn how a model asks your code to run a function and uses the result, so you can design tools the model calls correctly and safely. | 2026-10-21 |
| Transformer Architecture | Understand how attention lets a transformer weigh every token against every other, so you can read model specs and papers without getting lost. | 2026-10-21 |
| What Is Machine Learning? | Machine learning is writing programs whose behaviour is learned from examples instead of spelled out in rules; this page explains the idea, the three main kinds of learning, how training and testing work, and how today's chatbots fit into the same picture. | 2026-10-21 |