Before EliseAI was a $2.2 billion company, it was a hunch about a headache. In 2017, Minna Song took a job as an administrative assistant at a residential real estate firm in New York - not to climb the org chart, but to watch, up close, how leasing offices actually worked. What she found was a grind of repetitive messages: the same tour requests, the same maintenance complaints, the same rent questions, arriving faster than any team could answer. So Song and her Cambridge classmate Tony Stoyanov, an engineer and machine-learning specialist, built software to answer them.
That software - first called MeetElise, later EliseAI - does something deliberately unglamorous. It reads and replies to the flood of prospect and resident communication that flows into a property, across email, text, web chat and, increasingly, the telephone. It schedules tours, chases applications, logs maintenance tickets, nudges renewals and helps collect rent. The pitch is not that AI is magical. The pitch is that a leasing office should never have to say "sorry for the wait."
"AI that improves how we live."
— EliseAI company tagline
The bet worked because the problem was bigger than any one landlord. EliseAI now says its assistant is used by roughly 70% of the 50 largest rental housing operators and owners in the United States - a customer list reported to include 28 of the top 30, with names like Greystar and Equity Residential. The company crossed $100 million in annual recurring revenue and, in August 2025, raised a $250 million Series E led by Andreessen Horowitz that doubled its valuation to about $2.2 billion.
