AI Transformation · Asset Intelligence · 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.
Real estate is my current proving ground. The architecture travels.
Real estate is my current proving ground. The architecture travels.
Selected Work
From zero AI capability to institutional intelligence
Systems Architecture
Unified Asset Management Platform
₹926 Cr Mixed-Use Development
Every product type had its own spreadsheet. Every analyst had their own version. When someone left, the model left with them. We're replacing that with a single financial architecture — one model that serves all product types, validates itself, generates MIS automatically, and eventually pulls live data from SAP. Co-built with the Head of Asset Management through intensive workshops. Designed so the business owns it completely.
1
Single model replacing fragmented project-level spreadsheets
3
Product types unified under one architecture
Applied AI
Leasing Intelligence Pipeline
WTC Thane — India's First Design District
What happens when AI sits inside the leasing cycle instead of outside it? Positioning decks that write themselves for specific tenant profiles. Walkthrough videos generated for individual prospects. Client collaterals that adapt to who's reading them. We built the system that takes market research in one end and produces ready-to-send materials out the other — from geo-intelligence and tenant targeting through to CRM pipeline tracking.
6-phase
Pipeline from research to lead generation
3
Phases shipped, 4th in progress
Organisation Building
Asset Intelligence Department
Welspun One Logistics Parks · 2025–Present
A logistics and real estate company with zero AI capability. No team, no mandate, no infrastructure. We started with a hackathon at IIT Bombay, hired the founding team, and are building an intelligence layer across the entire asset lifecycle — investment analysis, leasing, construction, and operations. Four verticals, each tied to a business outcome, each co-owned with the department head who lives with it daily.
4
Intelligence verticals launched
8+
Departments engaged across the portfolio
Data Architecture
Investment Intelligence Pipeline
WTC Thane · Worli · Devanahalli
Before any asset gets a leasing strategy or a construction plan, someone needs to answer: what does the market actually look like? We're building the system that answers that — scraping government databases, running demand models, mapping supply pipelines, and consolidating it into a research framework that any project team can pick up and use. Three assets are running through it now.
3
Major assets in pipeline
1
Standardised BD research framework in use across new evaluations
Method
How I build — and why it sticks
Every engagement follows the same loop.
Research
I don't start with technology. I start in the room where the work actually happens. Sit with the analysts building the models. Watch the leasing team assemble pitch decks. Map the real workflow, not the one on the process document.
Create
Build the system in live workshops with the people who'll own it. Not a handoff from consultant to client — a co-build where domain expertise and AI capability merge in real time. The business shapes the system because the business has to trust it.
Train
Transfer ownership completely. The goal isn't a system that needs me. It's a system that makes me unnecessary. Person-dependent knowledge becomes organisation-owned infrastructure.
I started at Welspun One with zero team, zero mandate, and a blank page. We ran a hackathon at IIT Bombay to find raw talent, hired the founding team, and are building the full Asset Intelligence function from scratch — financial models, a data lake backbone, BIM intelligence, leasing pipelines. Everything is co-created with business heads, designed for handoff from day one. That's the method: build for independence, not dependence.
How I Work
Three ways in.
1
Leadership Training for AI
For the MD who knows AI matters but hasn't found a way to make it part of how they actually think and decide. Intensive 1:1 or small-group sessions. We move from "we should be doing something with AI" to "this is now how I evaluate deals, run reviews, and challenge my team's assumptions."
This is how my longest engagement began — structured sessions with a managing director that became a trusted partnership and the foundation for a full transformation.
2
Virtual Chief AI Officer
You have a team. You might even have pilots running. What you don't have is someone who can see across all of them and tell you what's working, what's theatre, and what to kill. I become your external architecture layer — quarterly strategy, system design reviews, vendor assessment, and the honest voice that your internal team can't always be.
Ongoing oversight without the overhead of a full-time CAIO.
3
Full Transformation
Blank canvas. No existing AI capability. Leadership wants an intelligence operating system, not a collection of pilots. I build the department, hire the team, design the architecture, co-create with your business heads, and hand off a function that runs without me.
Exactly what I'm building at Welspun One — from zero to a four-vertical intelligence department with its own team, budget, and mandate.
All engagements begin with a short conversation to see if there's a fit.
Approach
How I think about transformation
01
Start with the asset lifecycle
Every asset-heavy 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 works 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 layer, unified models, institutional memory. When I leave, the intelligence stays. Person-dependent knowledge becomes organisation-owned infrastructure.
Thinking
Selected writing
Let's talk about what intelligence looks like
inside your business.
I work with companies making the shift from intuition-driven to intelligence-driven operations. Not a pilot. Not a proof of concept. Infrastructure that compounds. If that's where you're headed, I'd like to hear about it.