In partnership with
From Italy to a Nasdaq Reservation
How do you follow record-setting success? Get stronger. Take Pacaso. Their real estate co-ownership tech set records in Paris and London in 2024. No surprise. Coldwell Banker says 40% of wealthy Americans plan to buy abroad within a year. So adding 10+ new international destinations, including three in Italy, is big. They even reserved the Nasdaq ticker PCSO.
Paid advertisement for Pacaso’s Regulation A offering. Read the offering circular at invest.pacaso.com. Reserving a ticker symbol is not a guarantee that the company will go public. Listing on the NASDAQ is subject to approvals.
Hi everyone! I’m diving into something new: building my very first AI agent and I want to take you along for the ride. This is my first brain dump in what will hopefully become a whole series of posts, where I share everything I learn, stumble on, and figure out as I go.
My aim? To build in public and be open about the process, the good, the confusing, and the “aha!” moments. If you’re also thinking of getting started with AI agents or just curious how it all unfolds, you’re in the right place. As I keep building, these articles should grow into a living guide for new starters like me. Feel free to follow along, ask questions, and even share your own tips or struggles - I’m all ears!
Step 1: Decide What Your Agent Should Do
Start simple. Ask yourself: What do I want my AI agent to help with? It could be sorting files on your computer, helping customers, planning travel, or even brainstorming product ideas. The clearer the goal, the easier everything else gets.
Here is a general flow of how the block from top to flow should look like.
Step 2: Pick the Right Tool for You
Here are some great options for first-timers and pros alike:
No-code/Low-code Platforms:
n8n: Drag-and-drop workflows make connecting AI and other services super easy.
MindStudio: Hundreds of templates and a visual builder—no coding required.
Modular Frameworks (for people comfortable with coding):
LangChain: Perfect if you want to wire up custom steps, memory, and integrations with full control.
Botpress: Visual flow editor, built-in NLU and deployment channels for support, onboarding, and more.
AutoGen by Microsoft: Lets you automate creating agents with minimal coding, especially if you like Microsoft’s ecosystem.
Semantic Kernel: Best for integrating AI smoothly into existing apps or enterprise workflows.
Most of these tools offer free versions, so you can experiment without spending a dime! There might be many more but I am not going to spend a month on tools research. Got some tools you prefer to work with?
Step 3: A Sample Case: Build an Agent to Sort Your Files
Let’s make this practical: Say you want an AI agent that helps organize your hard drive.
Here’s how you’d set it up:
Define the Goal: “Help me organize my hard drive by document type.”
Agent’s Thinking:
Analyze your files.
Decide sorting rules (by document type, date, etc.).
Move files to the right folders.
Planning: Lay out each step, what inputs the agent needs, and what outputs it will create.
Build in Tool:
If you use n8n, connect nodes for file analysis, logic, and movement.
With LangChain, set up a chain for each decision and file action.
Execute and Evaluate: Test your agent! Let it organize a small folder first. See what it does, does it get everything right? Refine its instructions as needed
Tools I have chosen are:
n8n oon my desktop (see the heading screenshot)
Google Sheets
Real-World Magic
AI agents can take on bigger jobs, too! Companies use them to develop products, manage customer queries, sort massive datasets, and even book travel deals: all by following the simple cycle of goal-setting, planning, executing, and refining.
So lets start building! Let me know how it goes, and don’t hesitate to drop questions as you build, help is always just a message away!
MEET THE WRITER
Amod Mulay
I've progressed from a Java developer to leadership roles, gaining expertise in software architecture and product management. I'm passionate about creating solutions, meeting people, and driving innovation. In my free time, I enjoy drawing, biking, and trail running. Feel free to contact me for collaboration.
1 The views expressed here are Amod’s own and do not necessarily reflect those of his current or former employers.