I ship AI systems that replace manual work with automated decisions
Product leader with a track record at Stanza Living and Power Finance Corporation. I've cut 12-hour review cycles to 15 minutes, collapsed multi-day financial workflows to hours, and put real-time intelligence in front of 300+ ops staff daily.
See My Work in Action
A walkthrough of MoodBite — a voice-first AI food concierge I designed and shipped end-to-end.
Systems I've Shipped
Production AI that moved real metrics — not slide decks.
CV Validation Pipeline
Cut candidate review time from 12 hours to 15 minutes across 22 clusters
Problem
HR teams across 22 coliving clusters were manually reviewing CVs — a 12-hour per-batch process that created hiring bottlenecks and inconsistent quality standards.
Solution
Built an LLM-powered validation pipeline with structured extraction, scoring rubrics, and automated routing. Integrated directly into existing HRMS workflows.
Impact
Credit Memo Automation
Collapsed monthly credit memo preparation from days to hours on a ₹40,000 Cr portfolio
Problem
Credit analysts spent days each month manually compiling memos across a ₹40,000 Cr loan portfolio. Legacy data silos made entity resolution near-impossible.
Solution
Designed NL-driven reporting automations with entity resolution and risk scoring models. Built executive dashboards providing real-time portfolio health for the first time.
Impact
Real-Time Ops Intelligence
Eliminated T+2 day reporting lag with AI-powered voice, video, and ops tooling
Problem
Operations teams across multiple cities relied on manual reporting with a 2-day lag. No real-time visibility into occupancy, attendance, housekeeping, or support metrics.
Solution
Shipped LLM-powered ops tooling for voice, video, attendance, and housekeeping. Built real-time intelligence surfaces used daily by 300+ city-level staff.
Impact
My Approach to AI Products
Every system I build is designed around three principles.
Evaluation
Rigorous benchmarking using RAGAS, custom gold sets, and LLM-as-a-judge patterns. I don't ship until I can measure.
Guardrails
Content filters, PII detection, and adversarial testing baked in from day one. Trust isn't a feature — it's architecture.
Probabilistic Success
Systems that degrade gracefully and manage output variability. When LLMs are wrong, my systems still work.
Tools I've Built
End-to-end products I designed, built, and shipped independently.
MoodBite
Voice-first AI food concierge that understands mood, craving, and context to recommend exactly the right meal.
View Project
FeatureFit
AI-powered feature prioritization tool for product managers. Upload your backlog, get RICE-scored recommendations.
View Project
LLM Tuner
Prompt experimentation playground for testing, comparing, and iterating on LLM outputs side-by-side.
View ProjectStrategic Writing
AI Products Fail on Power, Not Models
Why the biggest risk in AI product development isn't model quality — it's organizational power dynamics.
LLM-EO: The Future of Search
How LLM Engine Optimization is reshaping the way content gets discovered and consumed online.
Designing for LLM-EO
A practical framework for making your product visible to LLM-powered search engines.
Looking for an AI Product Leader who ships?
I'm open to senior AI PM roles (50-60 LPA), advisory engagements, and select consulting projects. Let's talk about what I can build for your team.