# Building with LLMs

> Resources for building with LLMs. Anthropic and OpenAI prompt engineering guides, building effective agents, Model Context Protocol, Claude Code, OpenCode and GitHub Copilot, and running models locally.

Source: https://ali.id.au/practices/building/llms/ (last edited 2026-10-12)

## Prompting and patterns

* [Anthropic: prompt engineering](https://docs.claude.com/docs/build-with-claude/prompt-engineering/overview)
* [OpenAI: prompt engineering](https://platform.openai.com/docs/guides/prompt-engineering)
* [Prompt Engineering Guide](https://www.promptingguide.ai/)
* [Building effective agents](https://www.anthropic.com/engineering/building-effective-agents) - read before picking a framework

## Tools and protocols

* [Model Context Protocol](https://modelcontextprotocol.io/) - see also the [Dataverse MCP server](/platforms/power-platform-d365/development/)
* [Claude Code](https://code.claude.com/docs) and [best practices](https://www.anthropic.com/engineering/claude-code-best-practices)
* [gstack](https://github.com/garrytan/gstack) - Garry Tan's opinionated Claude Code setup
* [OpenCode](https://opencode.ai/) - open source, any model
* [Hermes Agent](https://hermes-agent.nousresearch.com/) - open source, from Nous Research
* [CrewAI](https://www.crewai.com/) - multi-agent framework in Python
* [Groq](https://console.groq.com/) - fast inference API, with a playground
* Agent shell commands failing on Windows with `fork ... 0xC0000142`? [Fix Git Bash under Mandatory ASLR](https://gist.github.com/alirobe/e79fadfe47c064514e4a8af5b1b9fb6b)
* [Anthropic skills](https://github.com/anthropics/skills)
* [ElevenLabs](https://elevenlabs.io/) - voice, and [D-ID](https://www.d-id.com/) - avatar video

## Running models locally

* [Ollama](https://ollama.com/)
* [LM Studio](https://lmstudio.ai/) - with a GUI
* [AI Dev Gallery](https://github.com/microsoft/ai-dev-gallery) - Microsoft's Windows app of on-device samples; works nicely with Phi and Intel silicon
* [Foundry Local](https://learn.microsoft.com/azure/foundry-local/) - Microsoft's on-device runtime

## Microsoft

* [GitHub Copilot docs](https://docs.github.com/copilot) and [Copilot in VS Code](https://code.visualstudio.com/docs/copilot/overview)
* [Copilot Studio](/platforms/power-platform-d365/copilot-studio/) - agents on the Power Platform
* [Microsoft 365 Copilot documentation](https://learn.microsoft.com/copilot/microsoft-365/)