Make Actual Money with AI: The Agentic Engineer

At a recent channel trade show, I heard several channel partners asking one another, “Do you do AI yet?”
This soon raised a question for me: “What does that mean?” What does it mean to “do AI?” What exactly do you actually do to “do AI?” It’s not like you’re selling seats, licenses or subscriptions; there’s nothing to install or integrate. So what does an MSP do to “do AI?”

In December, I posted an article titled “New Agentic MSP Opportunities Emerge & Explode” in my Substack, “The Agentic MSP,” which focused on the large number of tools emerging that can be used to build your own AI practice.
This got me thinking about when the cloud was new. Many MSPs were confused about whether they should build their own cloud centers on their premises or if they should broker or buy and resell someone else’s cloud services. As AWS, Azure and Google hit the scene, that question was pretty well answered.

The Age of AI Presents a Different Opportunity

When cloud became popular, major system sales of servers, storage and related products became almost obsolete. Many resellers weren’t sure how to replace that lost revenue. Some considered transitioning their business into application development, but the skills required for AppDev would be very expensive to acquire. A large portion transformed into managed service providers (MSPs).
Now that AI is taking center stage, many are confused about how to incorporate it into their offerings. As I asked, “What does one do to ‘do AI’?” More precisely, how do MSPs make actual money with AI?
As I explored these questions, the obvious answer emerged that we would need to head back to ‘Broker, Buy or Build.” I actually first wrote about this last September in “AI Brings Us Back to Broker, Buy, or Build” in The Agentic MSP Substack. However, as AI adjusts rapidly, much has changed since then.
You may remember that, just over a year ago, we were talking about citizen developers. These are people with knowledge of their own business processes who use low-code/no-code (LCNC) tiled interfaces to assemble the software needed to support those processes by simply moving tiles around the screen. We often talked about how this took a substantial part of the load off software developers’ plates, freeing them for higher-level development tasks.
Since then, citizen developers are now, by and large, completing their transition from the LCNC platforms to using popular AI platforms like ChatGPT, Gemini, Claude and others to create much of what they need. This ranges from simple business functions like writing emails, producing reports, creating marketing materials, simplifying workflows and much more.
It seems there's just not much opportunity for MSPs in that level of AI utilization -- but, please, read on!

Popular AI Misconceptions

Many people think the model functioning as a chatbot is the entirety of how people use AI. They know they can ask questions and get answers. Some refer to AI as “search on steroids,” mainly because that’s all they’re aware of.
Much has been written and said about the fear many other people have of AI. As it becomes increasingly autonomous, there’s a danger it could take over the world and wipe out the human race. You may find this hard to take seriously, but among those warning about this is Geoffrey Hinton, widely considered “The Godfather of AI.”

The AI Continuum

Perhaps the most generic misconception is that AI is a single thing existing at a single level. Actually, AI can be said to exist at several levels across a broad continuum. These levels include:

• Generative – Users instruct their AI platform to create various creative content, including text, images, videos, music and more.
• Assistive – The AI, usually a chatbot, assists the user in performing various tasks, performing research, summarizing large documents, and writing various documents.
• Collaborative or Copilot – The AI, still starting with a chatbot, uses tools to create various kinds of content in collaboration with the user. One popular example is using Anthropic's Claude to create new Microsoft Excel spreadsheets. Claude is so adept at this that Microsoft decided to change the large language model (LLM) underlying Copilot for Microsoft 365 from OpenAI’s ChatGPT to Claude.
• Autonomous – Using a dedicated interface or an integrated development environment (IDE), the user instructs an AI Agent to perform various tasks that may require multiple steps, including examination, evaluation, decision-making, planning and taking action to make necessary changes in the real world. Basic autonomous AI agents often interact repeatedly with the human-in-the-loop for further information and decision-support.
• Fully Autonomous – The role of the human-in-the-loop is solely to review results and approve or make change orders. The fully autonomous AI agent continues cycling, awaiting specific trigger events to resume its operation.

The Autonomy Paradox

When people like Godfather Hinton talk about their fear of an AI takeover and human demise, they are referring to these fully autonomous agents. The fear is that the lack of human supervision will leave these agents free to commit all manner of horrors.
The paradox is that, according to industry studies, fewer than 1% of current deployments feature fully autonomous AI agents. In other words, the thing we’re most afraid of is something nobody can find good use cases for.

The AI Skills Gap

As developers scale upward from chatbots to developing agents, they find they have people who are capable of “chatting” with AI, but what they really need are engineers who can architect and build the AI-based systems their company needs. Citizen Developers needed Software Developers for more complex projects, and they still do need someone in that role. Thus, the AI question in most companies is now changing from, "How do we get our people to use AI?" to, "Who is architecting the agents that will run our operations?"
These Agentic AI Systems bring several requirements, including:

• Perception & Grounding (The "Eyes and Ears")
• A true agentic system doesn't just wait for text input; it actively perceives its environment. Using Multimodal Input, it can ingest not just text, but images such as screenshots of errors, files in PDF, CSVs, MD and other formats, and system logs.
• The AI understands where it is. It knows, "I am operating in the production environment," or "I am looking at the Finance SharePoint." It does not hallucinate that it is in a void; rather, it is "grounded" in your specific business reality. This is referred to as Environment Grounding.
• Planning & Reasoning: The single biggest differentiator from a standard chatbot is that the system must be able to "think before it acts."

• Task Decomposition: The capability to take a vague goal and break it down into a step-by-step plan.
• Chain of Thought (CoT): The system must generate a hidden "internal monologue" where it justifies its decisions before executing them.
• Self-Correction: If a step fails, the system must possess the logic to retry or try a different method, rather than crashing or asking the user for help immediately.

• Tool Use & Action (The "Hands")
• An agentic system must be able to execute action, not just describe it. The system connects to the Model Context Protocol (MCP) to read and write data. It doesn't just say "I sent the email"; it actually calls the email API to send it. It also has a "menu" of tools such as a calculator, calendar, CRM, and email and the intelligence to select the ideal tool for the right moment without human intervention.
• To assure safety, it includes "human-in-the-loop" checkpoints for high-stakes actions such as approving refunds or agreements.
• Memory & State Management
• Standard LLMs generally have amnesia. That is, they forget everything once the chat window closes. Agentic systems must remember.
• In the short term, it remembers where it is in a multi-step workflow. For longer term persistence in uses retrieval augmented generation (RAG) to store knowledge in a vector database so it can recall facts from months ago. It also logs its own past successes and failures to learn what worked previously.
• Critique & Reflection

Reliable agents have a built-in quality control layer. Before showing an answer to the user, a secondary internal process reviews it to assure it fully answered the user’s question and presented a response in the appropriate format. It also has hard-coded rules that override the model if it tries to do something it is forbidden to do such as deleting data or sharing private information.

The Agentic Loop

A system is only "agentic" if it can autonomously loop through this cycle:
Perceive → Plan → Act (Tool Use) → Reflect (Did it work?) → Remember.

The Agentic Engineer

This need for Agentic AI Systems creates a necessary new role with a new skill set-- the Agentic Engineer.
Here’s the broker, buy, or build proposition. To take full advantage of the substantial revenue available from building complex AI systems, you may decide to try to train your people on the skills I’m going to describe next, or you can seek to hire people who already have those skills.
But these people can be expensive, and the training is extensive so it’s going to take plenty of time and money to get yourself there.
If that won't suit you, you can opt to partner with or subcontract Agentic Engineers who are rapidly emerging in the marketplace. Just as with other partnering, these professionals are happy to join you on sales calls, talk to your customer to learn what their needs are and then to fulfill those needs. You get to impress your customers without spending a dime up front. Isn’t collaboration just great?

When do you need an Agentic Engineer?

There are several ways you can find opportunities among your existing customer base to engage an Agentic Engineer, whether for fun or profit. As with any other project, you need to learn what your customer is doing and what challenges they are facing. Three characteristics of project opportunities for Agentic Engineering include:

• Complexity: The task requires more than three steps of logic (e.g., "Read email, look up sender in Salesforce, check inventory, then reply"). Citizen developers struggle to make this reliable.​
• Connectivity: The solution requires writing data back to a system of record. Giving a chatbot "write access" without an engineer's security architecture (MCP) is a massive risk.​
• Consistency: You need the solution to work 99% of the time, not 80%. "Prompt engineering" alone hits a ceiling; "System engineering" is required to break through it.

When you believe you may be in the presence of such an opportunity, it’s time to reach out to your friendly neighborhood Agentic Engineer, partner up and invite them to join you on a sales call. They will ask the questions required to determine if the opportunity you think you have is feasible. They’ll then help you scope it, price it, propose It and close on the project.
As you go through your vetting process, remember that the critical first step once you’ve decided on a partner is to enter into a mutually beneficial documented agreement.

Vetting Agentic Engineers

As with all external skill sets, it’s important for you to carefully vet your candidates. Here are some aspects to look out for. This is a specialized technical role responsible for architecting and assembling AI systems. Unlike a "normal user" who chats with a chatbot, the Agentic Engineer must:

• Orchestrate Context: They meticulously engineer what data is fed into the model to ensure accuracy, reduce hallucinations and moderate operating expense.
• Build "Scaffolding": They build the code and logic around the LLM. This includes connecting RAG pipelines, defining tools via MCP and managing multi-step workflows where one AI agent passes tasks to another.
• Move from "Chat" to "System": They transition from asking questions in chat to designing repeatable, reliable systems through engineering.
• Advance Context Engineering: Design and optimize extensive system prompts and context injection strategies to ensure high-fidelity model responses.
• Build RAG Pipeline Architecture: Build and maintain RAG systems, ensuring the AI can accurately retrieve and synthesize proprietary data.
• Implement Agentic Tooling & MCP: Implement MCP to connect LLMs securely to internal APIs, databases and third-party software, enabling the AI read and write data rather than just passively retrieving info.
• Evaluate reliability: Create automated test suites, called "evals", to measure the accuracy and reliability of AI outputs before deployment thus moving from more casual "vibe coding" to engineered reliability.

The role of the Agentic Engineer is still actively being defined, so the terminology is still undetermined. One common term applied to this role is AI Orchestrator. This emphasizes the skill of managing multiple models, APIs and data sources to work in harmony.
The Agentic Engineer is responsible for designing, building and refining complex AI workflows that go beyond simple chatbots. This role bridges the gap between business needs and raw AI capabilities by engineering the context, connectivity and control required for enterprise-grade solutions. The Agentic Engineer you engage will need the following skills:

• Fluency in "Mid-Code": Proficiency in Python or TypeScript to write the "glue code" that connects LLMs to the real-world using frameworks like LangChain, AutoGen or raw API calls.
• Orchestration Logic: Ability to design multi-agent flows where a "Planner Agent" delegates tasks to "Worker Agents".
• Security & Governance: Understanding of prompt injection risks and data privacy boundaries when connecting AI to business data.

The New Essential Roler

The Agentic Engineer is the bridge between the raw potential of Frontier Models like ChatGPT, Gemini or Claude and the specific reality of your business data. They are the builders who take perception, planning, action and memory and weave them into software that doesn't just chat but is able to actually do work.​
For assistance in locating potential Agentic Engineer partners, feel free to reach out to me at [email protected].

Posted by Howard M. Cohen on February 11, 2026