Wang Jin: From Microsoft Engineer to Real Estate AI Tech Founder
Interviewer & Editor: Xiao He
Self-Introduction
Wang Jin:
A quick intro about myself:
I used to be a software engineer at Microsoft. Now I’m a founder working on ADU Pilot — a platform that helps homeowners and professionals plan ADUs (Accessory Dwelling Units) and analyze related regulations. I’m also the host of a podcast called “Georgia Xiaoshuai” — I’ve been running it for 4–5 years, and it has achieved some pretty good outcomes.
Xiao He:
Yes! I really like your podcast. 10k+ subscribers and commercial sponsorships — that’s impressive!
Microsoft Experience ｜ From “Retirement Company” to the AI Boom
Xiao He:
You joined Microsoft before GPT went mainstream. Looking back, how would you describe that experience?
Wang Jin:
Right. When I joined, people still saw Microsoft as a “retirement-friendly company.”
At first, I worked like 30-something hours a week. But within six months, it jumped to 50–60 hours — reflecting the rapid shift in the industry.
I was at Microsoft from 2021 to 2025 — exactly the time AI exploded.
In the beginning, I barely understood foundational models like BERT or LSTM. But in real work, productivity gains from AI were very obvious.
Why Entrepreneurship? — Sparked by Podcast Guests
Wang Jin:
Many podcast guests were entrepreneurs. Even if not “Silicon Valley founders,” they had built something from 0 to 1 — opened stores, built brands, ran traditional businesses.
My first job was at a startup of a dozen people that later got acquired. Then I worked many years in Big Tech — but never felt fully fit with that pace.
By year 3 or 4, I thought:
If my next job isn’t a startup, then I should start one myself.
Plus, I love real estate investing. I talk a lot with realtors and investors, and naturally start thinking:
What can AI do in real estate?
First Startup Idea: AI-Recognizing Home Photos
Wang Jin:
My realtor friends complained about two pain points:
• Is the “vaulted ceiling” really vaulted?
→ Listings often call 10-ft ceilings vaulted when they’re not.
• Is “waterfront” really waterfront?
→ A neighborhood lake ≠ your backyard faces water.
So I let AI read listing photos and identify specific property features.
Accuracy started at 70%. With prompt-design and workflow improvements, I pushed it close to 99%. But:
✘ Realtors weren’t willing to pay
✘ Demand wasn’t truly essential
✘ Hard to scale
Lesson:
If customers won’t pay, it’s not real value.
So I pivoted quickly.
Inspiration: An 11-Month Legal Battle
Wang Jin:
I met a homeowner who bought a house and wanted to build an ADU — but his permit got stuck. He felt the city rules were unreasonable and spent 11 months suing. He eventually won — but wasted huge time and money.
And I thought:
Why can’t we evaluate ADU feasibility before buying a home?
Why is information so opaque?
So I built ADU Pilot.
What Is ADU Pilot？—— Professional ADU Feasibility Reports
We integrate information including:
• State laws, city zoning rules, building codes
• Setbacks, privacy, solar/shadow rules
• Parking requirements
• Wildfire / flood risks
• Historical district constraints
• Slope / foundation risk
• Airport noise zone
• Timeline + cost estimation
• ROI and rental yield analysis
We simulate different structure sizes:
For example: 749 sqft vs. 1000 sqft —
How cost and ROI differ?
Payback period? Rental income?
Goal:
Before planning or construction, homeowners & architects already know what’s possible.
Market Insight: High Interest Rates Make ADUs More Attractive
Xiao He:
With high interest rates, people might not sell homes — maybe they renovate and add units instead.
Wang Jin:
Exactly! Like the Lipstick Effect in economics:
Can’t buy a new home → invest in the one you have
California also offers generous incentives for more housing:
• SB9 — Split a lot into up to 4 homes
• Density bonuses
• Various fee reductions
📌 The homeowner market is far bigger than the buying-and-selling market.
Business Model: Start with Professionals
Target customers:
We’re currently focused on B2B (architects, realtors),
but homeowners are the long-term growth driver.
Team Size
Wang Jin:
Fewer than 5 core members + several part-time contributors.
Podcast Recommendation: Georgia Xiaoshuai
Wang Jin:
Please check out my podcast “Georgia Xiaoshuai”. I interview first-generation Chinese immigrants in North America — mostly in Chinese — to tell real stories of immigration.
Xiao He:
Sharing those stories is so meaningful. I’ve thought about podcasting, but editing takes a lot of time — so I started with written interviews first. I respect how you persist, meet so many people, and keep delivering value.
Wang Jin:
Thanks!
I think the key is positive feedback loops.
Persistence itself is neutral — you only continue when you feel rewarded.
Bonus: Is an $8,000 Conference Booth Worth It?
Wang Jin:
I’m debating whether to pay $8,000 for a booth at an industry conference.
Xiao He:
My thoughts:
✔ Go to the conference and build trust face-to-face
✘ Don’t buy the booth yet (high cost & unclear ROI)
Low-cost alternatives:
• Partner with real-estate meetups
• LinkedIn / X / YouTube / Reddit / BiggerPockets
• Demonstrate reports using real addresses
• Host your own mini-trainings
• Publish content consistently — let your brand speak
$8,000 can fuel many high-quality exposures.
Wang Jin:
That’s excellent advice!
Closing Thoughts
In a fragmented, opaque, and asymmetric housing-information world,
Wang Jin & ADU Pilot are making it possible for ordinary people to make clear and confident housing decisions.
That is a value that technology should create.