I spent twelve years inside banking, wealth management, and cross-border strategy across India and Southeast Asia — managing portfolios, navigating regulators, and learning what risk actually looks like. Now I build the governance and trust infrastructure that enterprise AI is missing.
Most enterprises can deploy AI. Few can tell you which agents they have, what those agents decide, or whether anyone is governing them.
Enterprise AI without a registry is shadow IT. I build governance infrastructure - agent identity, decision audit trails, and autonomy guardrails - so organizations know what their AI is doing and why.
Models degrade silently. I design monitoring systems for drift, fairness, and hallucination - calibrated for India's data patterns, where monsoon cycles and festival spending look like anomalies but aren't.
Four regulators, overlapping jurisdictions, evolving frameworks. I build compliance layers that map AI decisions to regulatory requirements - audit-ready trails, explainability reports, and cross-regulation conflict resolution.
Risk frameworks designed around UPI, Account Aggregator, and GST data flows - not bolted onto legacy systems. Continuous credit assessment, signal-driven alerts, and adaptive models with built-in governance.
Redesigned the enterprise risk architecture for one of India's largest private banks. Unified credit, market, and operational risk under a single taxonomy. Deployed ML-based credit scoring and scenario stress testing aligned with RBI's revised risk weight guidelines. Basel III compliant six months early, releasing $320M in capital.
Built a multi-dimensional credit risk framework for a $650M commercial vehicle portfolio. Segment-specific models incorporating macroeconomic correlation and geospatial risk mapping - calibrated for India's CV cycle, GST-driven fleet formalization, and infrastructure capex patterns.
Led a Fortune 500 institution's expansion across five Southeast Asian jurisdictions during rapid India-ASEAN corridor growth (bilateral trade crossing $130B). Regulatory licensing, corporate structuring, tax optimization reducing the effective rate by 8.5pp. $450M capital deployed, $85M released through optimization.
Most banks still run models calibrated on pre-UPI, pre-Account Aggregator data. Whoever rebuilds the risk stack around new data flows will define the next cycle.
It will do for lending what UPI did for payments - but only if lenders build the ML infrastructure to use the data. The bottleneck is analytical capability, not regulation.
India's wealth management market is growing at 15%+ CAGR but advisory tooling is stuck in 2015. Goals-based advisory, tax-aware rebalancing, real-time portfolio intelligence - table stakes globally, greenfield here.
The firms that figure out the regulatory arbitrage early will have a structural advantage that compounds.
Indian enterprises are scaling AI agent deployments without central registries, reasoning capture, or policy enforcement. Agent sprawl is the new shadow IT. The governance infrastructure doesn't exist yet - it needs to be built.
RBI's AI risk frameworks, SEBI's algorithmic advisory guidelines, IRDAI's automation rules, and the DPDP Act are converging. Enterprises that build governance now will have a structural advantage. Those that wait will retrofit under pressure.
Over 20% of 2025’s $5B+ megadeals had an AI thesis at their core. DiligenceSquared just launched voice agents for M&A research. Deloitte has published extensively...
India has accidentally built the world’s best alternative data infrastructure for equity factor investing. Not deliberately – nobody at NPCI or GSTN was thinking about...
It’s been roughly a year since I started building AI systems full-time for financial services. Before that, I spent over a decade on the other...
Let's talk.
Download Resume (PDF)