AI Transformation · Real Estate · Systems Architecture
I build intelligence systems inside companies that have never had them.
Most organizations hire AI consultants who leave behind slide decks. I embed inside the business, build alongside domain experts, and hand off systems that outlast my involvement.
Thesis
The hardest part of AI transformation isn't the technology — it's rewiring how an organization thinks. I specialize in companies making the leap from intuition-driven to intelligence-driven operations. Not by replacing people, but by making them dramatically better at what they already do.
Selected Work
From zero AI capability to institutional intelligence
Enterprise AI Transformation
Asset Intelligence Department
Welspun One Logistics Parks · 2025–Present
Built an AI function from scratch inside India's fastest-growing logistics & real estate platform. Designed and implemented intelligence systems spanning investment analysis, leasing, construction (BIM), and facility operations across a multi-asset portfolio.
75%
Time reduction in financial modeling
1000+
Files processed across departments
Market Intelligence System
Investment & Leasing Intelligence Pipeline
WTC Thane · Worli · Devanahalli
Designed an end-to-end market intelligence system combining government database scraping, ML-driven demand modeling, geo-intelligence, and CRM pipeline analytics. Created a standardized BD research framework now used across all new project evaluations.
3
Major assets covered
7-phase
Research → GTM pipeline
Financial Model Architecture
Unified Asset Management Platform
₹926 Cr Mixed-Use Development
Co-built a reusable financial modeling system with the Head of Asset Management. One model serving all product types, with built-in validation, automated MIS, and a pathway to SAP integration. Designed for handoff — the business runs it, not us.
1–2
Analyst roles optimized
3
Product types unified
Approach
How I think about transformation
01
Start with the asset lifecycle
Every real estate business follows the same arc: invest, construct, lease, operate, exit. Intelligence should map to this lifecycle, not to an org chart. I work backwards from business outcomes, not forward from technology capabilities.
02
Build alongside, not above
The only AI implementations that survive are the ones built with domain experts, not handed to them. I sit in the workshops. I learn the domain. The handoff is seamless because the business was co-authoring from day one.
03
Systems over solutions
I don't build tools — I build systems that compound. A shared data lake, unified models, institutional memory. When I leave, the intelligence stays. Person-dependent knowledge becomes organization-owned infrastructure.
Thinking
Selected writing
Let's talk about
what you're building.
I work with companies that are serious about embedding intelligence into their operations — not as a pilot, but as infrastructure. If that's where you're headed, I'd like to hear about it.