The lab notebook
Field notes on AI engineering
824 articles · page 22 of 28
Voice AI Agents: Architecture and Latency Challenges
A practical breakdown of voice AI agent architecture, the latency budget behind natural conversation, and where most pipelines break.
What Employers Actually Look for in Junior AI Agent Engineers
A hiring-manager view of the skills, portfolio signals, and interview habits that actually separate hired junior agent engineers from ignored resumes.
Common Excuses That Stop People From Finishing Agent Bootcamps
The real reasons people abandon agent bootcamps have less to do with time and more to do with fear, ego, and bad habits.
Writing an Effective CLAUDE.md File: A Complete Guide
Learn how to write a CLAUDE.md file that actually changes how Claude Code behaves, with real examples and common mistakes to avoid.
From Zero to Agent Engineer: A Realistic 90-Day Plan
A no-fluff, week-by-week 90-day roadmap for becoming a working agent engineer, from Python basics to shipping a production agent.
Automating Ragas Reports for Stakeholder Reviews
Turn raw Ragas evaluation scores into scheduled, stakeholder-ready reports with automated pipelines instead of manual spreadsheet wrangling.
Combining Ragas with Human Review for Higher-Stakes RAG Apps
Learn why automated Ragas metrics alone aren't enough for high-stakes RAG systems and how to design a human review layer that catches what automation misses.
Common Ragas Setup Errors and How to Resolve Them
A field guide to the most common Ragas installation, API key, async, and schema errors, with the exact fixes that get your evaluation pipeline running again.
LangChain Agents Deep Dive: ReAct, Tool-Calling and Custom Agents
A practical walkthrough of LangChain agents covering ReAct reasoning, native tool-calling, and how to build custom agent loops that actually hold up in production.
LangChain Async Support: Building High-Throughput Applications
A practical guide to LangChain's async APIs, ainvoke, abatch, and streaming, for building LLM apps that survive real production traffic.
LangChain Best Practices for Production Deployments
A practical guide to LangChain best practices for production: structure, error handling, observability, cost control, and testing.
LangChain Caching: Reducing Redundant LLM Calls
Learn how LangChain caching cuts LLM costs and latency by avoiding redundant calls, with practical patterns for in-memory, SQLite, and Redis backends.
LangChain Callbacks: Hooking Into Every Step of a Chain
Learn how LangChain callbacks let you observe, log, and debug every LLM call, tool run, and chain step in real time.
LangChain Chat Models vs LLMs: Understanding the Interface Difference
A practical breakdown of LangChain's ChatModel and LLM interfaces, why they behave differently, and which one you should reach for in production.
LangChain Community vs Core: Understanding the Package Split
A practical breakdown of langchain-core vs langchain-community, why the split happened, and how to structure imports so your app stays stable.
LangChain Contextual Compression: Trimming Retrieved Documents
Learn how LangChain's ContextualCompressionRetriever trims irrelevant text from retrieved documents to cut token costs and sharpen RAG answers.
LangChain Cost Tracking: Monitoring Token Usage Across Chains
Learn how to track token usage and API spend across LangChain chains using callbacks, custom handlers, and per-user cost attribution.
LangChain Custom Tools: Writing Your Own Tool Class
Learn how to build production-ready LangChain custom tools using the @tool decorator and BaseTool subclass, with real code and error-handling patterns.
LangChain Debugging: Using Verbose Mode and Tracing Effectively
Learn how to debug LangChain apps fast using verbose mode, callbacks, and tracing so you stop guessing why your chains fail.
LangChain Deployment: Serving Chains as an API
A practical guide to deploying LangChain chains as production APIs with LangServe and FastAPI, covering streaming, auth, and scaling.
LangChain Document Loaders: Ingesting PDFs, Web Pages and More
A practical guide to LangChain document loaders for PDFs, web pages, CSVs, and directories, with code for building a clean RAG ingestion pipeline.
LangChain for Document Q&A: A Complete Beginner Walkthrough
A hands-on walkthrough of building a LangChain document Q&A pipeline, from loading files to retrieval and answer generation.
LangChain Error Handling: Retries, Fallbacks and Timeouts
Learn how to build resilient LangChain apps with .with_retry(), .with_fallbacks(), and timeout controls that survive real-world API failures.
LangChain Expression Language (LCEL) Explained
A practical, code-first guide to LangChain Expression Language (LCEL) — the pipe operator, Runnables, streaming, and when to graduate to LangGraph.
LangChain Few-Shot Prompting: Dynamic Example Selection
Learn how LangChain's ExampleSelector picks the best few-shot examples at runtime instead of hardcoding a static prompt.
LangChain History: How the Framework Evolved to LCEL and LangGraph
A practical walkthrough of LangChain's evolution from a simple chaining library to LCEL and LangGraph, and what it means for builders today.
LangChain for Multi-Agent Systems: Composing Several Chains
Learn how to compose multiple LangChain chains into a coordinated multi-agent system with routing, shared state, and real code.
LangChain for Multi-Tenant SaaS Applications
A practical guide to building isolated, secure, and scalable LangChain pipelines for multi-tenant SaaS products.
LangChain Multi-Vector Retriever: Handling Summaries and Full Docs
Learn how LangChain's multi-vector retriever separates what you embed from what you return, fixing retrieval on long documents, tables, and images.
LangChain Output Parsers: Getting Structured Data from LLMs
Learn how LangChain output parsers turn messy LLM text into reliable structured data using PydanticOutputParser and StructuredOutputParser.