About AI Field Guide
47 practical guides on using AI in real work — built by someone actually doing the work.
AI Field Guide is 43 plain-English guides on using Claude, ChatGPT, Perplexity, Cursor, and Generative Engine Optimization (GEO) — each one built for a specific role at a specific skill level. Written by Mark Reeves for solo operators and small teams who want frameworks they can run today, not theory.
Why this exists
I tried 50 AI tools. Watched the YouTube guru economy. Realized nobody was writing the thing I needed.
Mark Reeves is a pen name. The real background: I'm a cofounder of a design and manufacturing company that builds certification-level equipment used globally in product certifications. I've been using AI tools in real work — not demos, not side projects — since the GPT-3 days. Some of what we've built using AI and LLMs: an LLM-based ISO 9000 QA system for manufacturing, covering receipt inspections through final QA; production software using Claude, ChatGPT, and Cursor, including programs deployed globally; a QA and troubleshooting application that monitors live deployments, diagnoses failures, and writes the fix code itself. We use these tools every day to run a real business — and every one of those things was built using the tools we describe in these guides.
Getting there meant figuring most of it out ourselves. Every time we went looking for a guide that matched where we actually were — specific tool, specific role, specific level — we'd open a YouTube video promising "the only AI prompt you'll ever need" and close it three minutes in. We'd open a 200-page AI book and skip the first nine chapters of theory looking for the first relevant topic. Often there wasn't one.
The gap was obvious: there's a flood of AI content, but very little of it is practical, role-specific, opinionated, and updated. The hype people don't update. The textbook people aren't shipping. Make no mistake — I'm not this verbose. Claude, Cursor, and ChatGPT assembled the sentences. We brought the knowledge, the real-world context, and the red pen.
So we started documenting what we learned — first as internal training material, the founding documents we used to get new hires up to speed on what AI actually is and how to use it in real work. Those documents are what these guides are. One became three. Three became thirty. Each one built around a specific tool, a specific role, a specific level — because that's how you actually train people.
Who Mark Reeves is
Not a Stanford professor. Not an "AI thought leader." A person who ships work and got tired of guessing.
The portrait is AI-generated — our best guess at what Mark looks like. The byline and face are for consistency across 47 guides and counting. If you want to reach the human behind it: hello@ai-field-guide.com — that's a person, not a help desk.
AI-generated portrait.
What makes this different
Four things that aren't in most AI books.
Role-based, not tool-based
We don't write "ChatGPT for everyone." We write ChatGPT for Marketers (Beginner) and a separate ChatGPT for Marketers (Professional). Different work, different problems, different guide.
Frameworks, not prompt dumps
Anyone can paste a prompt. Frameworks are what compound: Brand Voice Documents, prompt anatomy, campaign architectures, evaluation rubrics. The 5th and 50th time you use a guide it's still earning.
Updated as tools change
Lifetime updates aren't a marketing line — they're an obligation. When Claude ships a new mode or ChatGPT changes a tier, the affected guides update within 30 days. Direct buyers get the new version automatically.
Honest about limits
Every guide has a "what this tool can't do" section. We tell you what to not use the tool for. That section is often more valuable than the prompt list.
What we cover
Five tools. Eight roles. Three skill levels.
The AI Field Guide series spans the tools that actually matter for real work in 2026: Claude (Anthropic), ChatGPT (OpenAI), Perplexity (research and citation), Cursor (the AI-native code editor non-developers can also use), and Generative Engine Optimization (GEO) — the new discipline of getting cited by ChatGPT, Claude, Perplexity, and Google's AI Overviews.
Roles include: Business Owners, Solopreneurs, Marketers, Authors, Solo Clinicians, Non-Coders building software, and the merely Curious. Skill levels are split into Beginner (you've never used the tool), Intermediate (you've used it for 30+ days), and Professional (you're building production systems).
Browse all 47 guides or start with the free one — AI for the Curious.
Editorial standards
How a guide gets written.
Every AI Field Guide follows the same internal process: tested by the author in real work first, then written. No theory paragraph that isn't paired with a prompt or workflow you can run. Every claim about a tool's behavior is dated — because Claude 3.5 isn't Claude 4, and whatever ships next won't be either. The "what this tool can't do" section is non-negotiable on every guide.
We don't quote unnamed "AI experts" or repeat hype that hasn't been verified. If we say a tool can do something, we ran it. If we say a tool can't do something, we tried.
When we say a workflow runs, it has run — in production, on real data, with real consequences if it fails. That's the baseline.
For the clinician and author guides specifically: we've helped friends — doctors, lawyers, and writers — discover systems and processes that have streamlined their practices and helped finish books. We asked them to review the guides, particularly the sections covering IP, PII, and HIPAA considerations. Their input shaped how those sections are written. That said, every practice is different, every jurisdiction is different, and every situation carries its own variables. Your mileage will vary. Nothing in these guides is legal or medical advice — it's what worked in real contexts, documented honestly, with the appropriate limits stated clearly.
The lifetime updates promise
Buy a guide direct from us — get every future update free, forever.
When a tool changes meaningfully, the affected guides get rewritten within 30 days. Every direct buyer gets an email with the new PDF. No re-buying. No new edition shenanigans. This is the moat — the reason to buy direct over Amazon Kindle.
Kindle buyers: Amazon doesn't support the same delivery model, so Kindle copies show the version that was current at publish. If you want lifetime updates, buy direct.
Refund policy
30 days. No questions. Refunded manually.
If a guide doesn't deliver — wrong fit, wrong level, wrong tool, anything — email hello@ai-field-guide.com and you'll get a full refund. We'd rather refund a bad-fit buyer than have someone sit on a guide they'll never use. No "explain why" form. No friction.
Press & contact
For journalists, podcast bookers, and partners.
Email: hello@ai-field-guide.com
Standard description (copy-paste): AI Field Guide is a series of 47 practical guides on using Claude, ChatGPT, Perplexity, Cursor, and Generative Engine Optimization (GEO) in real work. Each guide is built around a specific role and a specific tool at a defined skill level. Written by Mark Reeves and published as PDFs at ai-field-guide.com and as Kindle eBooks on Amazon. The project exists to bridge the gap between AI hype and AI use — frameworks and workflows you can run today, kept current as the tools evolve.
What we're working on
Roadmap.
The 43-guide v1 series is live. Next on the list: deeper Cursor coverage for non-developers, additional GEO guides for service businesses, and expanded agent workflows. Tools we're watching closely: Claude Projects updates, ChatGPT Operator, the new generation of vertical AI agents, and how Google's AI Overviews change the search/discovery game.
This isn't a finished product dropped on Amazon and abandoned. It's alive.