Production-grade references for AI engineering teams
Checklists, playbooks, and cheatsheets distilled from real deployments. The same standards we apply internally, published for teams building production AI systems.
Choose the asset that matches your situation
These resources work best when they route a specific engineering decision, not when they sit in a generic library.
Unsure whether this should even be agentic?
Start with the enterprise assessment kit to score suitability, autonomy level, and governance risk before build effort compounds.
Open resource → PrioritizeMultiple enterprise AI initiatives are competing for budget
Use the portfolio triage worksheet to classify what should be funded, redesigned, simplified, or stopped before the roadmap hardens.
Open resource → ProcureComparing vendors, platforms, or advisory partners
Use the vendor evaluation scorecard when procurement and technical stakeholders need one architecture-based comparison frame.
Open resource → StabilizeAgent pilot works, but production feels fragile
Start with the production AI audit to review failure modes, state, observability, and rollout risk before scale exposes the gaps.
Open resource → BuildShipping LangGraph workflows with state and approvals
Use the LangGraph cheatsheet when the implementation problem is checkpointing, HITL gates, and state design.
Open resource → DebugRAG answers are wrong, weak, or unstable
Use the RAG quality checklist to walk the pipeline from ingestion and chunking through retrieval and final grounding.
Open resource → GovernAI governance needs a board-ready evidence package
Use the Board Evidence Package to brief executives, procurement, or audit committees with structured artifacts — maturity snapshot, governance map, kill list.
Open resource → MapProduction AI systems need a governance control map
Start with the Governance Control Map Sample to see how deployed systems are mapped to autonomy levels, permission violations flagged, and sovereignty gaps identified.
Open resource →The Production-Ready AI Agent Audit
50 Critical Checks Before You Go Live
Beyond the demo: autonomous AI agents fail in production for five predictable reasons — reliability, state management, tooling, observability, and security. This audit systematically checks each one.
View resource →AI Agent Engineering Playbook
Internal Production Standards
The engineering standards we apply to every agent system we build. 12 failure modes with mitigation patterns, production gates, and deployment checklists.
View resource →RAG Quality Checklist
Debugging Retrieval-Augmented Generation
Your RAG pipeline returns confident, well-formatted, completely wrong answers. This checklist walks through every failure point systematically — from document ingestion to final generation.
View resource →LangGraph State Management Cheatsheet
One-Page Reference
State schemas, checkpoint patterns, human-in-the-loop gates, and multi-agent coordination. Everything you need on a single page when building LangGraph applications.
View resource →Enterprise Agentic AI Assessment Kit
Decide Whether, When, and How to Deploy Autonomous AI
The same decision framework we use in Fortune 500 advisory engagements. Covers agentic suitability scoring, autonomy tier classification, tool permission risk assessment, and a 10-question "Is your use case agentic?" scorecard.
View resource →Agentic AI Vendor Evaluation Scorecard
Procurement-Ready Scoring For Advisory And Platform Decisions
A structured scorecard for comparing agentic AI vendors, advisory partners, and platform options. Focuses on architecture judgment, governance depth, production evidence, vendor neutrality, and capability transfer.
View resource →Enterprise AI Portfolio Triage Worksheet
Classify Which Initiatives To Fund, Hold, Redesign, Or Kill
Worksheet for enterprise AI leadership teams managing multiple pilots or business-unit initiatives. Maps maturity, autonomy, failure cost, and ownership so weak bets are stopped before they absorb more budget.
View resource →Governance Control Map — Sample
What a Production-Grade AI Governance Audit Looks Like
A redacted one-page sample of the Governance Control Map artifact produced in AW governance engagements. Shows how deployed AI systems are mapped to autonomy levels, permission violations flagged, sovereignty gaps identified, and an overall risk threshold scored.
View resource →Board Evidence Package for Enterprise AI
Six Artifacts Your CTO or VP Engineering Needs to Brief Leadership
A structured template package for enterprise AI champions who need to brief executives, procurement, or audit committees without soft change-management language. Designed for CTOs and VP Engineering presenting to boards, operating partners, or legal and compliance stakeholders.
View resource →Architecture Decision Record Kit for AI Systems
Document AI Design Choices Like a Principal Engineer
ADR template for AI/ML systems with 10 pre-filled examples covering the architecture decisions that matter most: single-agent vs multi-agent, RAG vs fine-tuning, checkpoint backends, vector DB selection, orchestration framework, and more.
View resource →From checklist to production
These resources cover what to check. We handle the engineering to get you there.
[ SUBMIT SPECS ]