AI & automation
How does an AI agent work in customer service?
How AI for customer service works: an agent that answers from company data (RAG), queries your systems, hands off to a person and is measured.
6 min readUpdated on
Guides to AI, software and e-commerce
Direct answers to the questions that come before a project: what it is, when to use it, what the risks are and how to start. Each guide cites its sources and points to the service that solves the problem.
Custom software
How do you show up in ChatGPT and other AI answers?
How to show up in ChatGPT, Google and Copilot answers: how AI assistants pick sources, indexing, crawlers, structured data, llms.txt and measurement.
6 min readUpdated on
AI & automation
What are Claude Code skills, and how should a team use them?
What Claude Code skills are, how on-demand loading saves context, when to use a skill, CLAUDE.md or a hook, and how a team tests and versions them.
6 min readUpdated on
AI & automation
What is graph engineering, and when should agents run as a graph?
What graph engineering is, how LangGraph models agents as stateful graphs, when a graph beats a free-running agent and the risks of that design.
6 min readUpdated on
AI & automation
What is harness engineering, and when does it matter to a business?
What harness engineering is, where the term came from, how it relates to context engineering and the agent loop, and when it matters to a business.
6 min readUpdated on
AI & automation
LLM cost in production: what drives the bill and how to control it
What drives LLM cost in production: input and output tokens, calls per task, model choice, context and caching, and how to forecast and track the bill.
6 min readUpdated on
AI & automation
When should you run an LLM locally instead of using an API?
When to run an LLM locally instead of calling an API: privacy and data residency, latency, cost drivers, hardware, quality trade-offs and hybrid routing.
7 min readUpdated on
AI & automation
What is loop engineering, and when should an AI agent run in a loop?
What loop engineering is, where the term came from, the kinds of AI agent loops, why the stop condition matters and the risks of agents running unattended.
6 min readUpdated on
AI & automation
What is an MCP server, and what is it for?
What an MCP server is, what it exposes to AI agents (tools, resources and prompts), how it compares with an API and how to run one safely in production.
6 min readUpdated on
E-commerce
What is a payment gateway, and how does integration work?
What a payment gateway is, how a transaction flows, the ways to integrate and their PCI DSS impact, and how to connect payments and refunds to the ERP.
7 min readUpdated on
Architecture, modernization and mentoring
How do you migrate from PHP 5 to PHP 8 without downtime?
How to migrate from PHP 5 to PHP 8 without downtime: unsupported versions, what breaks, tests before the change, step-by-step upgrades and a safe cutover.
5 min readUpdated on
AI & automation
What is RAG, and when should a business use it?
What RAG is, how it works (chunking, embeddings, hybrid search, reranking), when a business should use it, why it fails and how to measure answer quality.
6 min readUpdated on
AI & automation
What is Spec-Driven Development (SDD), and how does it work?
What Spec-Driven Development (SDD) is, how it works with AI coding agents, where Spec Kit and OpenSpec fit in, and when the method pays off for a team.
7 min readUpdated on
AI & automation
What is an AI agent, and when does a business need one?
What an AI agent is, how it works, how it differs from a chatbot and when a business should use one: the signals, the risks, the controls and how to start.
6 min readUpdated on
Tell us what you're building
Tell us the context: the problem, the timeline and what already exists. You talk directly to the engineer who will do the work.
Request a first call
Free first conversation · Reply within 1 business day