Domain Expertise Drives Impactful AI Workflow Creation

Key Quotes

Did you know that deep domain expertise--not just engineering chops--is now the critical skill for building powerful AI workflows? Kipp and Tina Huang, data scientist and YouTuber, dive into how anyone can use tools like ChatGPT and n8n to stop customers from leaving by automating and personalizing your workflows for true impact.

This quote highlights a significant shift in the AI landscape, according to Kipp and Tina Huang. They argue that deep knowledge within a specific field is becoming more valuable than purely technical engineering skills for creating effective AI workflows. This suggests that individuals with specialized expertise are best positioned to leverage AI tools to solve real-world problems.

So, the most useful workflows are usually not the coolest workflows. So that is very true. It's like, what's the coolest kind of workflow? You know, we can make all these robots. What's the most useful workflow? Uh, we can make reports. Exactly, right? Yeah. So, I'd say there's a lot of report workflows that we work on. These are very custom because companies generally have a very specific way of doing their reports, to their investors, to their stakeholders, you know, a lot of different types of people. So that's a very popular workflow that people try to automate with HTK AI.

Tina Huang explains that the most impactful AI applications are often practical and functional rather than flashy. She identifies report generation and customer service automation as highly popular and useful workflows that companies are implementing. Huang emphasizes that these workflows are valuable because they address specific, recurring business needs that companies have.

So, first thing is mapping out your own workflow, like your own human workflow of what you're doing. Generally speaking, there are probably things that you do which are quite repetitive, and it involves you having to do it many times. So, it's like a consistently repetitive task that is low risk. That's one of its important, you know, if it's a very high risk thing, you definitely don't want to be risking that. And something else I like to keep in mind is what AI is good at. So, for example, AI is very good at being available 24/7, is very good at consistency, being able to do the same things more or less, like agentic workflows, you can get them to be quite consistent, and personalization is a really big one.

Huang advises a structured approach to identifying opportunities for AI automation. She suggests that individuals should first map out their own repetitive tasks, prioritizing those that are low-risk. Huang also points out the key strengths of AI, such as 24/7 availability, consistency, and personalization, which should guide the selection of tasks suitable for automation.

The most common roadblocks is other people. It's data, and other people are the most common two. I have failed. True. I, I absolutely concur with that statement. Other people and and data. Oh, yeah, absolutely. I think, yeah, often times like when we do work with clients, literally like the first blocker that it comes across is other people because they're like, oh, like, you first have to convince stakeholders, right, that you want to do this. And it's often times like, with AI solutions, it's not actually as plug and play as people would like it to be.

Huang identifies human resistance and data challenges as the primary obstacles in implementing AI solutions. She notes that convincing stakeholders and integrating AI into existing, often incompatible, architectures are significant hurdles. Huang's experience indicates that the technical feasibility of an AI solution is frequently less of a barrier than organizational and data-related issues.

Well, we're thinking about building an agent. Before you actually just start and going and building it, the way I describe it, I analogy I like to use is, right, it's like a burger. You can have different kinds of ingredients in the burger. You need to have a bun, um, you need to have vegetables, hopefully, you need to have a patty, you need to have condiments. However, the type of patty, vegetable, bun can be different. You can have like a whole wheat bun, you can have different types of vegetables, you can have ketchup as opposed to mustard. But if you don't have all of these components, you don't really have a burger. You have like a piece of bread or like some weird sandwich situation. So, it's very similar to agents in this sense, because agents, there are certain characteristics that it does need to have to be considered an agentic workflow, or else you're going to have problems when you actually do implement it.

Huang uses a burger analogy to explain the fundamental components required for building an agentic workflow. She stresses that while the specific ingredients (like the type of bun or patty) can vary, certain core elements must be present for it to function as intended. Huang implies that omitting these essential components will result in a system that does not operate as a true agent.

Domain expertise. Oh, this is not actually AI-specific. Because in my opinion, the people who end up building the most valuable agentic systems that we've seen this, like, over and over again, are actually surprisingly not the engineers. They tend to be people who have very deep domain expertise in a field. Like, for example, we had some very interesting ones, like people in, like, pest control. You know, it's like, they just spent like 10 years doing pest control, or people who work in in a law firm, accounting firm, in like privacy for very specific niches, like drug development, pharmaceuticals. So, deep expertise is actually the most valuable thing of an agentic workflow, because that translates into your ability to actually evaluate the agent.

Huang reiterates the critical importance of domain expertise in building successful AI systems. She observes that individuals with extensive knowledge in a specific field, rather than just technical engineers, are often the most effective builders of valuable agentic workflows. Huang explains that this deep expertise is crucial for evaluating the performance and correctness of the AI agent.