Engineering blogs, worked end to end.

Five design skills for AI coding agents compared: what each skill does, what it needs, its risks, and which skill to use for each job.

A detailed comparison of Cohere Embed 5, Gemini Embedding 2, Voyage 4 and OpenAI text-embedding-3: pricing, limits, retrieval evidence, data handling and clear recommendations.

Compare the best open-weight and closed embedding models on licensing, retrieval evidence, cost, privacy and lock-in, with clear picks for RAG, search and on-device apps.

Build a RAG evaluation dataset, diagnose retrieval and answer failures, and tune accuracy, latency and cost around your customers' needs.

Learn how embeddings and vector search work, compare eight database options with their pros and cons, and choose the right one for RAG, search or recommendations.

The same RAG pipeline built in LangChain and in LlamaIndex, a side-by-side list of their prebuilt components, and which framework fits each common use case.

What drives the cost of building a custom RAG system, how to estimate monthly running costs from token volumes, and three worked scenarios with every assumption written down.

Lock down a Linux web server step by step: only ports 80 and 443 open to the internet, SSH reachable only over Tailscale, hardened sshd, automatic security updates and kernel settings, with commands for Ubuntu, Debian, RHEL, Rocky Linux, CentOS Stream and openSUSE.

Set up Gradio 6 and build AI web apps in Python: Interface and Blocks, streaming LLM chatbots, state, API and MCP access, authentication and deployment.

The two critical PHP remote code execution bugs fixed on 21 October 2026, explained in plain terms: how the attacks work, which servers are most at risk, what a break-in can cost, and how to patch supported and end-of-life PHP versions.

The same AI lead-intake workflow built in n8n and in custom Python, with the maintenance tradeoffs of each and when to choose which, or both.

How to evaluate an independent AI developer: a scoping checklist, questions to ask, what to check in the code at handover, an evaluation plan and a delivery process that protects you.