AI ENGINEERING
RAG Engineering / 01How RAG Works, from Documents to Answers
Understand how RAG connects documents to an LLM: build an index, retrieve passages, assemble context, and return an answer with traceable sources.
The notebook / 64 articles
Ideas into practice
Notes on LLMs, machine learning, data engineering, and the systems that make them work.
Showing 11–20 of 64 posts
AI ENGINEERING
RAG in 2026: From Vector Search Demo to Reliable Evidence System
A system-level guide to the twelve classic RAG failure modes, the techniques now used to address them, and the limits that still remain.
AI ENGINEERING
Standing on Open Source: How a Codex Edge Case Became Agentusage
How two OpenUsage projects inspired me to investigate a Codex JSONL edge case and build a local-first dashboard for AI coding agents.
DATA SCIENCE
Statistical Tests for Data Analysis: A Practical Onboarding Guide
Learn how to choose and use statistical tests without turning analysis into a p-value checklist—from experimental design and assumptions to effect sizes, confidence intervals, and...
AI ENGINEERING
Use Local Models in VS Code Copilot with LM Studio and Unsloth Studio
How to connect VS Code Copilot Chat to local OpenAI-compatible models served by LM Studio or Unsloth Studio, including the BYOK utility-model setting that prevents agent errors.
AI ENGINEERING
KV-Centric LLM Serving: vLLM, SGLang, and Disaggregated Attention
A practical mental model for KV cache memory, vLLM PagedAttention, SGLang RadixAttention, and prefill/decode disaggregation in modern LLM serving.
AI ENGINEERING
Reproducing vLLM and LMCache KV Cache Reuse on a CPU-Only MacBook
A reproducible CPU-only vLLM and LMCache workspace for understanding KV cache reuse, transfer modes, and local inference debugging on Apple Silicon.
AI ENGINEERING
Use Databricks Models with VS Code Copilot and Copilot CLI
How to use one local proxy to connect Databricks Model Serving to both VS Code Copilot Custom Endpoint and GitHub Copilot CLI.
AI ENGINEERING
A Git-Native Message Channel for Local Coding Agents
How I use a Git ref as a local message channel between Claude Code and Codex for implementation, independent review, and iterative bug fixing.
LARGE LANGUAGE MODELS
Attention Dilution
Why longer LLM context windows can make answers worse: softmax dilution, positional bias, and attention sinks.