The lab notebook
Field notes on AI engineering
824 articles · page 12 of 28
How to Build Your First MCP Server in TypeScript (Step-by-Step)
A hands-on guide to build an MCP server in TypeScript, from tool schema to Claude Desktop config, with the pitfalls that trip up most first attempts.
Claude Code Hooks and Skills: Automating Your Dev Workflow
A practical guide to Claude Code hooks and skills, with real config examples for automating linting, testing, guardrails, and reusable workflows.
Claude Code Subagents Explained: When and How to Use Them
A practical guide to Claude Code subagents — what they are, why context isolation matters, and how to use them without slowing yourself down.
Getting Started with Claude Code: A Beginner's Complete Guide
A hands-on beginner's guide to installing, prompting, and working with Claude Code without drowning in hype or jargon.
Claude Code vs Cursor vs GitHub Copilot: The 2026 Comparison
A hands-on 2026 comparison of Claude Code, Cursor, and GitHub Copilot — workflow, multi-file changes, pricing, and which one to actually pick.
How AI Coding Agents Handle Multi-File Refactors (With Examples)
A technical walkthrough of how AI coding agents plan, execute, and verify safe multi-file refactors across large codebases.
Codex vs Cursor vs Windsurf: Picking Your AI Coding Agent in 2026
Codex, Cursor, and Windsurf all write code well — the real decision is workflow fit, team setup, and how much IDE you actually want.
Integrating DeepEval into GitHub Actions: A CI/CD Recipe
A practical recipe for running DeepEval evals on every pull request, with caching, secrets, and PR comments.
DeepEval Metrics Explained: G-Eval, Faithfulness, Hallucination and More
A practitioner's guide to every major DeepEval metric — what each one measures, when to use it, and where teams misread the score.
DeepEval Tutorial: Unit Testing Your LLM App Like Pytest
A hands-on DeepEval tutorial showing how to write pytest-style unit tests for LLM outputs, complete with metrics, thresholds, and CI.
DeepEval vs Ragas vs LangSmith vs Braintrust: 2026 Comparison
A practitioner's breakdown of DeepEval, Ragas, LangSmith, and Braintrust to help you pick the right LLM eval stack in 2026.
DeepEval vs Ragas: Which LLM Eval Framework Should You Choose?
A practitioner's comparison of DeepEval and Ragas for LLM evaluation, covering setup, RAG metric depth, and when to migrate.
How to Evaluate a RAG Pipeline: Faithfulness, Recall and Precision Explained
A practical guide to evaluating RAG pipelines using context precision, recall, faithfulness, and answer relevancy, with a CI-ready eval loop.
Top LangChain Alternatives: LlamaIndex, Haystack and DSPy
A practitioner's guide to LangChain alternatives — LlamaIndex, Haystack, and DSPy — and when to pick each one over LangChain.
LangChain Memory Explained: Buffers, Summaries and Vector Stores
A practitioner's guide to LangChain memory — buffer, window, summary, entity, and vector-store patterns — and how to pick one by cost and conversation length.
LangChain Tutorial 2026: Build Your First Chain and Agent
A beginner-friendly LangChain tutorial for 2026 covering chains, tools, and your first working agent.
LangChain vs LangGraph: The Difference and When to Use Each
LangGraph doesn't replace LangChain — it sits on top of it for stateful, cyclical agent orchestration. Here's when you need which.
LangChain vs LlamaIndex for RAG: A Hands-On 2026 Comparison
A practitioner's side-by-side of LangChain and LlamaIndex for RAG, covering defaults, ecosystem, and when each one actually wins.
LangFlow Custom Components: Building Your Own Node
A hands-on guide to writing, testing, and packaging LangFlow custom components when built-in nodes stop covering your production logic.
Deploying LangFlow Pipelines to Production
A hands-on guide to taking LangFlow pipelines from the visual builder to a secure, monitored, scalable production deployment.
LangFlow Tutorial: Visual LangChain Pipelines Without Code
A practical LangFlow tutorial showing how to build, wire, and export RAG pipelines visually before touching a line of Python.
LangFlow vs n8n: Which Visual AI Builder Should You Use?
LangFlow builds AI pipelines; n8n automates business workflows with AI steps bolted on. Here's how to pick.
LangGraph Checkpointing and Memory: Persisting Agent State
A technical deep-dive into LangGraph checkpointing: how state is captured, persisted, resumed, and time-traveled in production agents.
Human-in-the-Loop with LangGraph: Interrupts and Approvals Explained
A practitioner's guide to pausing LangGraph agents for human approval before risky actions, with real interrupt and resume code.
LangGraph Tutorial: Building Stateful Multi-Step Agents
A hands-on LangGraph tutorial covering state, nodes, edges, and cycles for building agents that loop, reason, and call tools.
LangGraph vs CrewAI vs AutoGen: Choosing Your Orchestration Framework
A practitioner's comparison of LangGraph, CrewAI, and AutoGen to help you pick the right multi-agent orchestration framework.
LangSmith Evaluations Explained: Datasets, Evaluators and Experiments
A practitioner's guide to LangSmith evaluations — datasets, evaluators, and experiments — for testing LLM apps like real software.
Setting Up LangSmith Alerts for Production LLM Monitoring
A practical guide to LangSmith production monitoring — the metrics that matter, alert thresholds, routing, and runbooks for LLM apps.
LangSmith Tutorial: Tracing and Debugging LLM Apps
A practical LangSmith tutorial for tracing, debugging, and evaluating LLM chains before they break in production.
LangSmith vs Langfuse vs Helicone: LLM Observability Compared
LangSmith vs Langfuse vs Helicone compared on tracing depth, self-hosting, framework lock-in, and cost tracking — with a practitioner's take on when each one wins.