Guides

Curated articles and resources on agentic design, forward deployment engineering, ML/AI management, and prompt engineering best practices.

Agentic Design

CORE: Why Product Discovery Is Broken

Part 1 of the CORE Discovery series. Introduces the four-phase methodology (Capture, Orient, Refine, Execute) and the problem CORE was built to solve: teams collapsing discovery into a single undifferentiated "talk to customers" step.

product-discoverycore-methodologyai-coaching

bryanortizRead →

Agentic Design

CORE: Architecture of an AI Coaching Platform

Part 2 of the CORE Discovery series. Deep dive into the provider abstraction pattern, the five specialized AI agents, cross-phase context flow, and the design decisions behind a portable Next.js + FastAPI application that runs on a laptop or on Azure.

architectureai-agentsprovider-abstractionfastapinextjs

bryanortizRead →

Agentic Design

CORE: From Zero to Running in Ten Minutes

Part 3 of the CORE Discovery series. A hands-on walkthrough covering local setup with Ollama, Azure configuration with Cosmos DB and Azure OpenAI, and running your first discovery session through all four phases.

setuplocal-developmentazureollama

bryanortizRead →

Agentic Design

CORE Discovery Framework (GitHub)

Full source, architecture diagrams, BRD, TRD, and ADR-001 (provider abstraction rationale) for the CORE Discovery Framework. Next.js 16 + FastAPI + five AI agents + Vertex integration, published as open source.

open-sourcecore-frameworkarchitecture

GitHubRead →

Agentic Design

theGuard: Getting Started

How to set up theGuard in your project. Covers installation, the init wizard, platform auto-detection, email verification setup, and configuring guards for your workflow.

ai-safetysetupmcp

bryanbotRead →

Agentic Design

theGuard: Platform Registry and Rules

Deep dive into theGuard platform registry: 18 platforms, 170+ rules with severity levels, impact categories, and remediation guidance. Covers AWS, Azure, GCP, Kubernetes, Docker, Terraform, and more.

ai-safetyplatformsrules

bryanbotRead →

Agentic Design

theGuard: MCP Proxy for Multi-Tool Protection

How to use theGuard MCP proxy to intercept tool calls from any AI coding agent. Covers auto-discovery, tool routing, the guard_exec shell tool, and integration with Cursor, Copilot, Cline, and Windsurf.

mcpproxyai-safety

bryanbotRead →

Agentic Design

Building Effective Agents

A comprehensive guide from Anthropic on building effective AI agents. Covers agent architectures, workflow patterns like prompt chaining and routing, and when to use autonomous agents vs. simpler workflows.

agentsarchitectureworkflows

AnthropicRead →

Agentic Design

LangGraph Multi-Agent Workflows

LangChain's guide to building multi-agent systems with LangGraph. Explores supervisor patterns, agent handoffs, and how to orchestrate complex workflows across multiple specialized agents.

langgraphmulti-agentorchestration

LangChainRead →

Page 1 of 3 (27 guides)