Knowledge Hub (33)

Azure Linux is Microsoft's open-source, security-hardened Linux distribution built for Azure and AKS. Here's what it is, where it came from, and how to try it.

One dotnet new command gives you a Blazor chat-with-your-data app — vector search, document ingestion, citations, and Azure OpenAI wired up on keyless auth.

A .NET console chat app in one Program.cs file — using Microsoft.Extensions.AI so the OpenAI model behind it is a swappable detail, not a lock-in.

Where does the conversation history live? That one decision shapes your agent's cost, privacy, portability, and UX. Here's how service-managed and client-managed storage differ — and how Microsoft Agent Framework abstracts both behind one programming model.

Build a full-stack TODO app — ASP.NET Core API, SQLite, and a Vite React front-end — all wired together and run with a single F5 by .NET Aspire. Step by step, from empty folder to running app.

Run a model on your own machine during development, swap to Azure OpenAI in production — same code. Here's how Aspire and Ollama make local AI a registration detail, not a rewrite.

Four libraries cover almost everything you'll do with AI in .NET. Here's how Microsoft.Extensions.AI, VectorData, the Agent Framework, and MCP fit together — without juggling vendor SDKs.

Stop memorising Copilot features and start with a real task from your backlog. Here's how I split work between chat and agentic flows across Visual Studio, VS Code, the CLI, and the cloud agent.

AI agents waste tokens on filler words you never read. Caveman is a Claude Code skill that rewires how your agent speaks — same technical depth, 65% fewer output tokens, works in Claude Code and GitHub Copilot.

Before I let any AI agent touch my codebase, I make it show me a plan. Here's how that one habit changed my workflow — and why the benefits go far deeper than 'fewer bugs'.

Move your CrewAI projects from local scripts to reliable services. Learn deployment architectures, observability patterns, and operational guardrails for production workloads.

Cut costs and latency in multi-agent workflows. Learn token budgeting, parallel task execution, caching strategies, and model tiering to make your CrewAI crews production-ready.

Trace what your agents actually did and why. Master CrewAI callbacks, event listeners, usage metrics, and log capture to diagnose failures before they reach production.

Give agents memory across runs. Master short-term context, long-term persistence, and state isolation to build workflows that learn and adapt.

Master tool design for CrewAI. Build custom tools, integrate APIs, compose tools, and optimize performance for production agents.

Build multi-agent systems with CrewAI. Learn core concepts, structure autonomous teams, and avoid common orchestration pitfalls.

Claude Code stores context in CLAUDE.md files and an auto-memory system. Here's the mental model and practical patterns for keeping it clean and useful across sessions.

MCP servers plug external tools into Claude Code — query your database, search Notion, browse Linear, control a browser. Here's how to add and configure them.

Claude Code subagents are independent AI workers for scoped tasks like search, planning, and security review. Here's when to use them and how to author your own.

Hooks in Claude Code fire shell commands at lifecycle events — before tool runs, after edits, on session start. The escape hatch for linting, formatting, and gating risky actions.

Claude Code is keyboard-first — the built-in shortcuts let you interrupt agents, paste multi-line prompts, scroll diffs, and switch modes. Full cheat sheet.

Slash commands are Claude Code's direct control surface for session state, memory management, and custom workflows. Full built-in reference plus how to author your own.

Claude Code is Anthropic's agentic CLI that reads your repo, edits files, and runs tests from the terminal. This guide covers installation, setup, and your first session.

Oh My Posh is a cross-platform prompt engine that shows Git status, active language versions, and command timing in your terminal. Here's how to set it up.

Winget is Microsoft's built-in Windows package manager — install, update, and manage thousands of apps from the terminal with a single command. Here's how.

Zed is a GPU-rendered code editor built in Rust by the Atom creators — millisecond startup, real-time collaboration, and built-in AI out of the box. Here's my take.

GitHub Copilot CLI brings agentic AI directly to your terminal — write code, run tests, and fix errors without context-switching. Here's how to set it up.

Hugo and Markdown are the quiet backbone of modern developer publishing — plain text in, blazing-fast static site out. Here's why this pairing works so well.

Hugo is the fastest static site generator available — no database, no runtime, just Markdown files compiled to blazing-fast HTML in milliseconds. Here's how it works.

Windows Terminal finally gives Windows developers a first-class command line — tabs, panes, GPU rendering, and WSL integration all in one place. Here's how to get the most out of it.

My Python pipeline worked on Windows but crashed on Linux in production. Dev Containers fixed the environment mismatch — here's how I set them up.

Struggling with messy console logs locally? I set up Serilog and Seq in my .NET 8 Web API for structured, searchable logs — here's the complete walkthrough.

Your AI outputs are only as good as your inputs. Here's how I use prompt engineering techniques to get consistently strong results from ChatGPT, Copilot, and image generators.