The wiki
Long-form guides on the ideas behind the news — architectures, agent patterns, and how this whole operation runs.
Loading wiki…
Practical techniques for cutting LLM API bills: prompt caching, model routing, batch APIs, and token discipline.
Oct 13, 2026 · 5 min read
API versus self-hosted LLM deployment, inference servers, latency and throughput basics, caching, and monitoring.
Oct 13, 2026 · 5 min read
How to run structured adversarial testing on LLM applications using the OWASP Top 10 for LLM Applications.
Oct 13, 2026 · 5 min read
A practical guide to measuring LLM output quality with metrics, LLM judges, and human evaluation.
Oct 13, 2026 · 5 min read
A factual comparison of LangGraph, CrewAI, AutoGen, and the OpenAI Agents SDK across control flow, state handling, human-in-the-loop support, and maturity.
Oct 13, 2026 · 7 min read
How language models emit structured tool calls, how to design tool schemas, and how to handle the failure modes that show up in production.
Oct 13, 2026 · 7 min read
A practical guide to building an AI agent around the plan act observe loop, covering tool definitions, stopping conditions, error handling, and trace logging.
Oct 13, 2026 · 7 min read
How low-rank adaptation trains a fraction of a model's parameters, with a minimal working peft example and QLoRA quantization.
Oct 13, 2026 · 7 min read
When retrieval-augmented generation is enough, when you need to update model weights, and how to choose between the two.
Oct 13, 2026 · 6 min read
How to measure a RAG pipeline: retrieval metrics, generation faithfulness, RAGAS-style LLM judges, human evaluation, and the failure modes to watch for.
Oct 13, 2026 · 5 min read
How to split documents into chunks for retrieval, from fixed-size and recursive splitting to structure-aware and parent-document strategies, with working code.
Oct 13, 2026 · 4 min read
A practical comparison of pgvector, Pinecone, Qdrant, Weaviate, Milvus, and Chroma for RAG workloads, across hosting, indexing, filtering, and licensing.
Oct 13, 2026 · 5 min read
How retrieval-augmented generation works: the indexing, retrieval, and generation stages, plus hybrid retrieval and advanced patterns.
Oct 13, 2026 · 4 min read
Getting better answers on multi-step problems by having the model reason step by step before answering.
Oct 13, 2026 · 4 min read
How to steer model output with a handful of input and output examples, without any fine-tuning.
Oct 13, 2026 · 4 min read
How to write system prompts that set role, rules, and output format for every request in a conversation.
Oct 13, 2026 · 4 min read
Practical techniques for writing prompts that get consistent, correct output from language models.
Oct 13, 2026 · 4 min read
Showing 17 of 63 articles.