Yesterday, I built my first AI agent in about an hour—with zero coding—using ChatKit and OpenAI’s Agent Builder, right after finishing this LinkedIn Learning course: Build Your First AI Assistant with OpenAI Agent Builder (No Code Required). Note - you need to have a premium LinkedIn subscription in order to access the course. I'm sure there are other courses out there, this is just the one that appealed to me the most!

I'm celebrating my first agent build in a small way; as you can see from the image, this accomplishment, while big to me, is really small potatoes when it comes to the complexity of building agents, with only one connector needed! However, doing this has increased my confidence in building on top of this learning experience.*** Does anyone know any other good courses they'd recommend? ***Drop it in the comments!

Wrangling my Inbox

I’m not a developer, and certainly don't know Python; I'm just a curious, tech-comfortable person who lives in email all day. My goal was simple: create an assistant that could read my emails, summarize them, and draft professional responses so I could spend less time stuck in my inbox and more time doing real work.

The course walks you step-by-step through the process, and that alone gave me the confidence to even try. Using Agent Builder felt a bit like filling out a series of smart forms rather than “programming.” I told the agent what I wanted it to do, gave it some guidance on my tone and style, and connected it to my email. Before long, I had an assistant that could scan messages, pull out what mattered, and suggest polished replies that still sounded like me.

**Here are the instructions I gave it, guided by the LinkedIn course: **

"You are an AI email management assistant that helps users stay organized and save time. Your job is to: read and summarize the key points of my email. Suggest a clear, well-structured draft response in a polite, professional, empathetic tone. If the message includes specific questions, ensure each one is answered. When summarizing, keep it concise - 2-3 bullet points; capture tone, intent, and any requests or deadlines. When drafting a reply, begin with a greeting, acknowledge the sender's message or request. Provide a clear response or next step. End with a friendly closing."

It wasn’t all smooth sailing, though. In the first video section of the course, the interface had changed just enough since the recording to confuse me. A few buttons and screens looked different than what I saw on my own screen. It wasn’t a huge change, but it was enough to stump me and slow me down while I tried to figure out where everything had moved. I also had to figure out how to run it in ChatGPT. If you’ve ever paused a video, squinted at your screen, and thought, “Wait… where is that menu on my version?”, you know the feeling.

The good news is: I did figure it out, and I took notes as I went. If you decide to take the same course and get stuck in those early steps because the interface doesn’t quite match, I’m happy to share the modified instructions I used. Just email me at [email protected] and I’ll send them to you. I love sharing what I’m learning so others don’t have to trip over the same hurdles.

I don’t claim to be an AI expert—far from it. I’m just someone who is learning out loud and experimenting in real time. If building an AI agent feels out of reach, I want you to know it’s more doable than it looks, especially with a no-code tool like Agent Builder and a guide like this course. And if my mix of mistakes and small wins can make your path a little easier, that feels like a success to me.

Like this post? [Follow me on LinkedIn!](https://Follow me on LinkedIn: www.linkedin.com/comm/mynetwork/discovery-see-all?usecase=PEOPLE_FOLLOWS&followMember=clada)