Companion source code for AI APIs in Practice, a visual engineering book on multimodal AI, agents, RAG, evaluation, governance, and production-ready products. Book DOI: 10.5281/zenodo.21431944 · Softw
A Visual AI API Playbook for Multimodal, Agentic, Production-Ready, and Commercial Applications
Run the executable showcase: Open in Google Colab · View saved outputs on GitHub
https://faramarzkowsari.github.io/ai-api-playbook/
AI APIs in Practice is an open research-software project by Faramarz Kowsari. It translates modern AI API capabilities into reproducible engineering patterns, offline-testable examples, safety utilities, evaluation workflows, and commercially realistic product blueprints.
The repository is organized by durable capabilities rather than rapidly changing model names. It covers language models, structured outputs, tool use, agents, RAG, multimodal systems, image generation, OCR, speech, realtime interaction, video, routing, evaluation, safety, observability, and unit economics.
The default configuration runs in deterministic mock mode and does not require paid API credentials.
Status: First public engineering edition, version 1.0.0. Provider availability, model names, prices, SDK interfaces, and regional access may change. Consult the provider matrix and each provider’s current official documentation before production deployment.
Resource Link ------ Executable Google Colab showcase Open and run Saved program outputs View Notebook Source-code examples Browse examples Commercial product blueprints Browse projects Architecture documentation Read architecture guide Provider coverage View provider matrix Automated test runs GitHub Actions Version 1.0.0 release GitHub Release Permanent scientific record Zenodo DOI Machine-readable citation CITATION.cff
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/FaramarzKowsari--ai-api-playbook.html)
Turn scattered notes, docs and transcripts into a queryable Markdown wiki — an LLM knowledge-base compiler with MCP access, no embeddings, self-hosted.
Azure-Samples/foundry-hosted-agents-workshopA hands-on, step-by-step workshop for building a Travel Assistant agent on Microsoft Foundry hosted agents with the Agent Framework - growing from a single agent to a multi-agent planner with function
ozankasikci/global-agent-memoryLocal-first, project-aware memory MCP server for Claude Code, Codex, and other AI agents, with Obsidian and an owner dashboard.
vinodborole/okf-kitTurn any website into a portable, agent-ready Open Knowledge Format (OKF) bundle — no LLM required. Crawl, sync, chat offline, and serve docs to coding agents over MCP.
AdamRussak/self-docSelf-hosted documentation RAG pipeline that serves semantic search to LLM agents over the Model Context Protocol.
KernelLord/pickysteveSkill router and context picker for Claude Code, Cursor, and Aider. Auto-picks the right skill for every prompt, with prompt-injection filtering for MCP.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.