In-depth pieces on LLM internals, retrieval, agents, MCP, and evals.
The latest pieces across every category, newest first.
Structural, recursive and semantic splitting compared, with the sizes to start from for short lookups and for prose. Chunk size sets a ceiling on what retrieval can return, and several common mistakes lower it further.
Listed inWhat are AI Agents?AI Agentson
Anthropic on the difference between workflows and agents, five composable patterns, and why most teams should start with the simplest thing that works.
Listed inTokensHow LLMs Workon
Character and word counts disagree with real token counts, which is a problem because every limit you hit is counted in tokens. How byte pair encoding splits text, and how to budget a request so a response is not cut off mid-sentence.
Prepending model-generated context to each chunk before embedding, and the measured drop in failed retrievals it produced.
Listed inData LayerModel Context Protocolon
The normative source: lifecycle, capability negotiation, transports, and the exact shape of tools, resources, and prompts.
Listed inModel Context Protocol (MCP)Model Context Protocolon
Hosts, clients and servers, and the three primitives a server exposes: tools, resources and prompts. Includes a working TypeScript server in about forty lines, and the mistakes that make a server unusable in practice.
Listed inModel Context Protocol (MCP)Model Context Protocolon
The announcement that started the ecosystem — why a shared protocol beats N×M bespoke integrations.
Listed inWhat is an AI Engineer?Introductionon
The day-to-day work is context plumbing, schema design, evaluation and cost control rather than model training. The prerequisite that carries most of the weight is ordinary software engineering.
Listed inMulti-Agent SystemsAI Agentson
Lilian Weng's survey of planning, memory, and tool use — still the clearest map of the agent design space.
Listed inWhat is an AI Engineer?Introductionon
A single support-reply feature, taken through the failures it hits in production: supplying a refund policy, grounding replies in help-center articles, structuring output for a ticket tool, and handling prompt injection in customer email. The role is the engineering around the model call.
Listed inGoogle GeminiModels & Providerson
Images, audio and video travel in the request and are billed as input tokens, at different per-second rates for audio and video. The same models are reachable through the Gemini Developer API and through Google Cloud, which differ in the platform around them rather than in capability.
Listed inLong Context ProcessingContext Engineeringon
The paper behind the U-shaped recall curve: models attend to the start and end of a long context and lose the middle.
Listed inPerforming Similarity SearchVector Databaseson
Cosine and dot product rank the same on normalized vectors. The decision that actually changes your results is where the filter runs.
Listed inPrompt Injection AttacksSafety & Ethicson
Years of worked examples showing why filtering does not solve injection, and why the dual-LLM pattern is the closest thing to a fix.
Listed inTransport LayerModel Context Protocolon
stdio or streamable HTTP, at revision 2026-07-28, which removed sessions, GET streams and resumability. What moving off stdio forces you to add.
Pinecone walks through fixed, recursive, and document-aware splitting with concrete size and overlap recommendations.
Listed inDeterministic EvalsEvaluationon
An evaluation is deterministic when it is computed rather than judged by a model, which is not the same as exact rather than fuzzy. A similarity score is computed, so it sits between exact matching and a judge model, and it costs nothing to run.
Listed inHow LLMs WorkHow LLMs Workon
The transformer paper. Worth reading once even if you never train anything — everything downstream inherits its shape.
Listed inInferenceIntroductionon
What happens between your request and the answer: time to first token against total response time, why output length drives latency, where token costs accumulate, and the inference failures to handle in code.
Listed inFunction CallingAPIs & SDKson
OpenAI's reference on declaring tools, handling the model's call, strict schemas, and parallel tool use.