Tech Pulse · AI & automation
AI & automation.
AI agents and n8n workflows: what to automate, how to build it safely, and what it costs.
Page 3 of 4
The Architect’s Blueprint: Building a Fully Autonomous AI Content Engine
In the age of generative AI, the digital landscape is being flooded with “grey goo”—generic, uninspired content produced by lazy, one-shot prompts. To stand out, businesses must transition from simple automation to Orchestrated Intelligence. At Goodish Agency, we’ve engineered a framework that…
Automating “Where Is My Order?” (WISMO) with n8n
Tired of endless “Where Is My Order?” inquiries? Proactive AI automation transforms support by identifying shipping issues and notifying customers before they ask. This reduces tickets, builds trust, and turns customer service into a powerful loyalty engine.
What Is an Agentic Workflow? Agentic vs Linear Automation (2026)
An agentic workflow gives an AI agent a goal and a set of tools and lets it choose the steps. Here is when that beats linear automation, and how to keep it safe.
Building Resilient AI: Implementing Retry Logic in n8n
Robust n8n error handling is more than a chore; it’s a critical ingredient for building AI automation you can trust. It transforms fragile workflows into professional-grade, ‘set and forget’ engines that inspire confidence and ensure system resilience.
Run Local LLMs in n8n with Ollama (Llama 3 and Newer): 2026 Guide
Install Ollama, pull a model, and point n8n's Ollama node at port 11434. Here is the setup we use, the Docker networking gotchas, and when a cloud model is still the better choice.
AI Memory in n8n: Short-Term vs Long-Term Memory for Agents
n8n agents forget between runs unless you give them memory. When to use Simple Memory, Postgres or Redis chat memory, or a vector store.
Building Self-Healing AI Workflows in n8n
Standard AI automation is fragile and costly to fix. The solution is self-healing AI: a system that spots its own mistakes, diagnoses the cause, and automatically applies a fix, transforming brittle processes into resilient, autonomous operations.
Multi-Agent Orchestration in n8n: Building AI Teams
Single prompts struggle with complex tasks. Multi-agent orchestration solves this by coordinating a team of specialized AIs, each an expert in its sub-task. Instead of one AI doing everything, you build a team to achieve sophisticated goals.
Human-in-the-loop AI agents for marketing ops: the review queue
Our rule for AI agents in marketing and sales ops: the agent prepares, a person approves, and only then does anything reach a customer or the CRM. Here is how the review queue works.
LangChain vs n8n for AI agents (2026): when to use which
LangChain is a code framework, n8n is a workflow platform. Here is how we decide between them for AI agents, and why many production setups end up using both.