OpenAI’s Dot Found Me a NYC Apartment. I Just Showed Up.
I gave dot a goal, made decisions along the way, and showed up. It helped with the apartment search, viewings, paperwork, and the little things after signing.
NYC apartment hunting feels like a competitive sport… except the prize is getting to pay rent.
My instructions were simple: find me a place close to work and schedule the viewings.
A few days later, I was opening Google Calendar and going wherever it told me.
If there’s a harder city to apartment-hunt in, I’m not volunteering to test it. Listings, availability, broker replies, applications, documents, follow-ups… apparently I needed a project manager to rent a studio.
So I used dot, OpenAI’s AI assistant.
It searched StreetEasy and other listings around my preferences and location goals. It helped compare commutes and gave me listing links alongside Google Maps links so I could quickly see where everything was.
Then it emailed brokers and leasing offices, worked through availability, and scheduled viewings around my calendar. When meetings changed, I could tell it to ask about moving a viewing instead of restarting the whole conversation myself.
The calendar became the interface. Where am I going? When? How do I get there? Who do I contact when I arrive? What does this place cost? Those were the details I wanted in the event, including transit time between stops.
That last part matters. Putting two apartments on a calendar ten minutes apart does not magically make the subway faster. Yet.
I opened my calendar and went to the next place. That was a pretty good user experience.
Once I picked an apartment, the work kept going. dot helped gather and organize the application documents, read through the lease, surfaced terms worth questioning, and handled emails with the leasing office. I could ask questions about the actual document instead of trying to keep every detail in my head.
Then dot caught something I hadn’t: Monday was a bank holiday. I needed a cashier’s check… on a Sunday. So it found open branches and worked out how to get me there.
Apparently my AI apartment assistant was now running bank errands. I was still the legs.
I still chose the place, signed the lease, and paid. Apparently AI has not solved the part where rent comes out of my account.
And after signing, there was another entire category of stuff to figure out. Getting keys. Move-in instructions. Building rules. What mattress fits the Murphy bed.
Yes… even confirming whether I needed a queen mattress became an email thread. Very advanced civilization we have here.
dot helped work through those questions and collected the relevant files in one place. It also helped prepare a little introduction for my neighbors. Small things individually. Quite a lot of things together.
There were corrections and a few rounds of ‘no, that’s not what I meant.’ It still needed me to pay attention. Welcome to managing literally anything.
But compared with personally managing every search, reply, appointment, document, and follow-up… I put in very little effort. I gave it a goal, made decisions along the way, and showed up.
That is a useful way to judge an AI assistant. Give it something in your life that involves a bunch of websites, conversations, and annoying little decisions. See how much of the process it can actually carry.
Finding an apartment gave mine plenty to do.
The future is here… it’s asking whether the mattress can be more than 10 inches thick. Lol 😆
