How I Bought a Car with Obsidian and ChatGPT—and Sold It Using a Spammer Filter

3 minute read

I moved to the U.S. in June 2021 for my PhD. By the end of 2023, I applied for one internship—and somehow, I got it.

But there was a catch.

The internship was in another city, and we needed a second car. I wasn’t exactly thrilled. I had only been driving for about 7–8 months and didn’t feel confident choosing a good used car.

To make it a bit more fun, we planned a New Year’s trip to Boston with friends. So I thought—why not try to buy a car during the trip?

📌 Step 1: Obsidian + Facebook + Too Many Cars

Early in the morning, while everyone was asleep, I opened Facebook Marketplace and started searching. I listed over 30 potential cars on our route from New York to Boston. Then I did something that changed is kind of unusual…

I dumped all the listings into an Obsidian note. Specs, prices, links, everything. One place. I thought, maybe if I organize it well, I can analyze and compare more easily.

But boy… it was overwhelming. I had no idea what I was really looking at. It felt like 30 browser tabs open in my brain.

🤖 Step 2: ChatGPT Enters the Chat

So I tried something fun. I copied the entire note and pasted it into ChatGPT. I asked everything I could think of—price comparisons, mileage insights, reliability issues, etc.

After a long conversation, ChatGPT gave me a shortlist of 5 cars worth calling for a test drive. I felt like I had a personal assistant. I coordinated with sellers while my wife and friend drove. We managed to test three cars on the way. The second one was it—a 2010 Toyota RAV4. Clean, smooth, and cheaper than my existing 2014 CR-V.

We sealed the deal. It served me perfectly for my internship. I loved driving it.

🚘 Step 3: Selling the Car—And the Spammer Nightmare

Fast forward to mid-2024. My wife moved to Tennessee. Therefore, we didn’t need two cars anymore, so we decided to sell the RAV4.

I quickly took a few pictures and posted them on Facebook.

Within hours, I received over 10 messages.

But none of them felt real—all spammers. Fake profiles, weird messages, sketchy offers. It was frustrating.

The car sat for days. Insurance was still active, and we didn’t want the extra liability. Meanwhile, I was still working full-time in my internship and had no idea what city I’d move to next.

💡 Step 4: The Engineer Mode

Next weekend, I cleaned the car properly, waited for perfect daylight, and took crisp pictures. Then I sat down with focus and created a new Obsidian note—this time like a pro seller. I crafted this message for the new post:

“I bought this car last December for my internship in Knoxville and have since driven it 8,000 miles without any issues. The car offers a smooth driving experience and is in good condition with a clean title. I’m selling it now because my internship is ending and I’m relocating. Highlights:

New tires and brakes

Reliable performance

Well-maintained and regularly serviced

Title is in my name, ready to transfer

Note: If you’re a scammer, don’t waste your time. I’m an engineer with a good knowledge of cybersecurity.”

To my surprise, the inbox exploded—but this time, all genuine buyers.

I set up a few meetings and sold the car the by next week—at a higher price than what I paid, after 8,000 miles of driving!

Sure, the money was nice, but what mattered more was the lesson.

🎯 The Hidden Lesson: Spammer Filter

As an engineer, I’ve always learned about filters—high-pass, low-pass, band-pass.

But this was the first time I experimented a spammer-pass filter—crafted not with circuits, but with words and intention.

Turns out, your Obsidian notes, a bit of ChatGPT magic, and a carefully written description can go a long way.

Would I recommend this method? Absolutely. Steal it when you need.

👋 About Me

Hi, I’m Shuvangkar Das, a power systems researcher with a Ph.D. in Electrical Engineering from Clarkson University. I work at the intersection of power electronics, DER, IBR, and AI — building greener, smarter, and more stable grids. Currently, I’m a Research Engineer at EPRI (though everything I share here reflects my personal experience, not my employer’s views).

Over the years, I’ve worked on real-world projects involving large scale EMT simulation and firmware development for grid-forming and grid following inverter and reinforcement learning (RL). I also publish technical content and share hands-on insights with the goal of making complex ideas accessible to engineers and researchers.

📺 Subscribe to my YouTube channel, where I share tutorials, code walk-throughs, and research productivity tips.

Connect with me: