AI Agents
17 articles · written by Niraj Kumar
An agent is a loop with tools, and almost every production problem with agents comes from that loop: it runs too long, calls the wrong tool, loses its state on restart, or takes an action nobody approved. Model quality is rarely the bottleneck.
These articles cover agent architecture from that angle — tool interface design, when a state machine beats a free-running loop, durable checkpointing so a crashed run can resume, human approval gates on destructive actions, and the specific pitfalls that show up once an agent handles real traffic.
Several posts pair agents with retrieval, since most useful agents are also RAG systems.
- Event-Driven ArchitectureLangGraphNode.js
Architecting Event-Driven AI Agent Workflows: Integrating LangGraph, Node.js, and AWS for Real-Time Automation
Stop polling APIs. Learn how to build real-time, event-driven multi-agent AI workflows with LangGraph, Ollama local LLMs, Node.js, Express, n8n, Make.com, Socket.io, Next.js, Docker, and AWS — a practical 2026 blueprint for sovereign AI automation.
- AI AgentsLangGraphDocker
Architecting Sandboxed Code-Execution Agents: Ephemeral Docker Runtimes, LangGraph, and Real-Time Telemetry in Next.js
Build a zero-trust, sandboxed AI coding agent with LangGraph, Ollama, and Claude Code that runs Python, C++, and JavaScript in ephemeral, network-restricted Docker containers on AWS Linux, with JWT execution policies, PostgreSQL audit trails, and live Socket.io telemetry in a Next.js workspace.
- MCPModel Context ProtocolLangGraph
Engineering Custom MCP Servers for Local AI Agents: Bridging PostgreSQL, MongoDB, and REST APIs with LangGraph and Next.js
A hands-on 2026 guide to building secure TypeScript MCP servers that connect local Ollama models and Claude Code to PostgreSQL, MongoDB, and REST APIs, orchestrated with LangGraph, streamed to a Next.js dashboard, and deployed with Docker and Nginx on AWS.
- LangGraphHuman-in-the-LoopAI Agents
Engineering Durable Human-in-the-Loop (HITL) AI Workflows: State Checkpointing with LangGraph, PostgreSQL, and Next.js
Build enterprise-grade human-in-the-loop AI agents that pause for approval before Stripe or Razorpay payouts. Learn LangGraph interrupts, PostgreSQL checkpointing, Socket.io real-time dashboards in Next.js, an Express resume API, JWT/OAuth RBAC, n8n alerts, and Docker deployment on AWS.
- LangChainAI AgentsNext.js
Automating Enterprise B2B SaaS: Integrating LangChain AI Agents, Next.js, and Zapier Workflows on AWS
A practical 2026 blueprint for building an AI-driven B2B SaaS: LangChain agents, a Next.js + Tailwind + Shadcn UI frontend, secure Express APIs with OAuth and JWT, Zapier automations, and zero-downtime Docker, Nginx and AWS deployments.
- PostgreSQLCDCSemantic Cache
How to Prevent AI Context Poisoning using PostgreSQL CDC for Real-Time Semantic Cache Invalidation in Node.js
Learn how to stop stale semantic caches from poisoning your AI agents. Build a real-time invalidation pipeline with PostgreSQL CDC, a Node.js worker, and an embedding similarity radius to flush outdated cache entries instantly.
- LangGraph.jsZodAI Agents
How to Build a Self-Healing Tool Call Execution Loop using Zod and LangGraph.js
Learn how to build a self-healing tool call loop in LangGraph.js. Validate LLM tool arguments with Zod, route failures to a correction node, and let local models fix their own invalid JSON.
- AI AgentsNode.jsVector Database
How to Build an Agentic Memory Garbage Collector in Node.js to Prevent Vector Store Context Poisoning
Learn how to build a Node.js memory garbage collector for AI agents. Implement time-weighted relevance decay, semantic deduplication, and safe pruning to stop vector store context poisoning and hallucination cascades.
- LangGraphPostgreSQLAI Agents
How to Implement Transactional State Rollbacks for AI Agents in LangGraph and PostgreSQL
Learn how to wrap LangGraph tool nodes in PostgreSQL savepoints, roll back partial failures deterministically, and stop duplicate billing before your LLM retries.
- LangGraph.jsAI AgentsContext Engineering
How to Implement Hierarchical Context Compaction in LangGraph.js to Prevent Agent Context Drift
A practical, code-first guide to building a hierarchical context compaction layer in LangGraph.js — using Ollama for local summarization and PostgreSQL checkpointing to stop long-running AI agents from drifting off-goal.
- DockerNode.jsAI Agents
How to Build a Secure Docker-Isolated Code Execution Sandbox for AI Agents in Node.js
Learn how to safely run untrusted LLM-generated code using Node.js and the Docker Engine API. A practical guide to ephemeral containers, cgroup limits, network isolation, and seccomp profiles for AI agent sandboxes.
- RPACrewAIPlaywright
How to Replace Legacy RPA with Autonomous Multi-Agent Workflows using CrewAI, Playwright, and Node.js
Learn how to replace brittle, rule-based RPA scripts with resilient, goal-driven multi-agent systems built with CrewAI, Playwright, and Node.js. Includes real code for autonomous web navigation, JSON validation, and PostgreSQL integration.
- Claude CodeGitHub ActionsCI/CD
How to Automate Multi-Axis PR Code Reviews using Claude Code and GitHub Actions CI/CD
Wire Claude Code into GitHub Actions to review pull requests across security, performance and testability — line-by-line, before a human opens the diff.
- LangGraphHuman-in-the-LoopPostgreSQL
How to Implement Human-in-the-Loop (HITL) Approval Gates in LangGraph with PostgreSQL Checkpointers
Build production human-in-the-loop approval gates in LangGraph with AsyncPostgresSaver, interrupt_before, and Command-based resume flows.
- LangGraphAI AgentsLangChain
Beyond Basic Chatbots: Developing Advanced AI Agents with LangGraph
Build stateful, multi-step AI agents with LangGraph — core concepts, real architectures, code examples, and the pitfalls beyond simple chatbots.
- AI AgentsLLMTool Calling
Designing AI Agents: Tools, Patterns, and Pitfalls
Designing AI agents that hold up — single-step vs multi-agent workflows, tool-calling patterns, memory architecture, and the reliability pitfalls.
- Graph-RAGKnowledge GraphsRAG
Graph‑RAG Explained: Building Smarter AI Agents with Knowledge Graphs
How Graph-RAG combines knowledge graphs with LLMs to improve multi-hop reasoning and accuracy — with code examples and enterprise use cases.
Need this built rather than explained?
Production LangChain & LangGraph systems for RAG, agents, and AI workflows.
LangChain Developer Services →Related topics
- Retrieval-Augmented Generation (RAG)
- LangChain & LangGraph
- Large Language Models
- Next.js
- React
- TypeScript
- Node.js
- PostgreSQL
- Docker & Containers
- DevOps
- CI/CD
- Application Security
- AWS
- n8n Automation