// Advisory work · Independent practice
Three engagements. Each one begins with a structured examination of the decision chain — not a technology review, not a process audit. A structural governance interrogation.
// What happens if you do nothing
The drift continues. Decisions accumulate. Each one individually defensible. The aggregate moving quietly away from what the board intended. No dashboard shows it.
A regulator asks a question. An audit finding recurs. A board member wants to understand a specific decision from eighteen months ago. Reconstruction begins. It takes weeks. The answer is incomplete.
The audit identifies every layer where the chain has broken. The Decision Integrity Blueprint™ gives the board one finding and one action. The drift stops being invisible.
Start the audit →// The practitioner behind the framework
25 years · FSI and healthcare · 25+ jurisdictions
Global banks. Insurers. Financial market infrastructure. Regulatory engagement across 25+ jurisdictions. The pattern that became the DIC™ framework was built from operating inside the environments where these failures actually happen — not from observing them.
Two named outcomes
Global bank liquidity transformation — revenue rank up 3 positions in two years. Global insurer continental digitisation — consistent governance architecture across multiple countries, languages, and regulatory environments.
Cases, domain cards, regulatory scope, and the research behind the framework: See the full body of work →
01 · 10-day engagement · Fixed fee
Map your institution's decision chain against the DIC™. Find every layer where intent, rules, and execution have diverged. Delivers the Decision Integrity Blueprint™ — not a 300-page report.
02 · Board session · Half day
A structured working session with your board or executive committee. Three questions. Three layers. Every one designed to surface what the dashboards cannot show. Not a presentation — a working session.
03 · Ongoing advisory · Monthly
Monthly engagement. Governance questions as they arise. Board papers reviewed. Decision architecture designed. An independent practitioner embedded in your governance cycle — not in your politics.
// Typical reasons institutions call
AI model approved. Outcomes are inconsistent across channels or customer segments. Nobody inside can explain the divergence.
A board member or regulator asks a question about a decision made 12–18 months ago. The room goes quiet. The answer takes weeks to reconstruct.
A supervisory review or audit is 60–90 days away. The institution cannot demonstrate that AI-driven decisions were within the boundaries the board approved.
The same governance gap surfaces in audit after audit. Teams respond. The finding returns. The root cause is architectural, not operational.
A large programme — digital transformation, platform migration, AI rollout — is two years in. The board's original intent and the current operating model no longer match.
Multiple systems are making decisions that affect customers. Nobody can identify who owns any given outcome if a regulator asks. The accountability chain is missing.
// Who this work is for
High-consequence institutions where the chain from board intent to operational outcome has grown longer, faster, and more automated than governance has kept up with.
"By the time a board sees decision drift, it has usually been operating for 18–36 months."
"The question is not whether the AI is right. The question is whether the institution can govern what it has built."
// Engagement on record — composite, anonymised
The board could not explain why lending outcomes differed across channels running the same approved model. Three credit decisioning sequences could not be reconstructed end-to-end from available records.
Two decision chains had no documented authority boundary at L5. One had a missing deployment authorisation record at L6. Authority boundaries redesigned. Board remediation initiated within 30 days of the finding.
Composite case based on documented engagement patterns. Institution not identified. All findings anonymised.
// Start the conversation
Every engagement begins the same way: one question about a decision your board cannot fully reconstruct. What happens next is structured, confidential, and specific to your institution.
A loan portfolio shift, a model deployment, a board-approved initiative — where the outcome does not match the intent. You want to understand exactly where the chain broke.
Not yet ready to name the specific decision. You want a confidential introductory conversation — under NDA — to understand what the audit involves and whether it applies to your institution.
// Start the conversation
Select a path above, then write to deepak@lumathink.com — all initial engagements are treated as confidential. An NDA can be executed before any substantive discussion.
Write to deepak@lumathink.com →Singapore-based. Working globally. Independent practice. All initial engagements treated as confidential.
deepak@lumathink.com · Singapore · Working globally
// Disclaimer & Legal Notices
The views, analyses, and perspectives expressed on this site are solely those of Deepak Aggarwal, presented in a personal and independent capacity. They do not represent or reflect the views, policies, or positions of any current or past employer, client, organisation, or affiliated entity.
All institutional case references — including but not limited to SVB, Knight Capital Group, Wirecard AG, Credit Suisse, Wells Fargo, Coutts, Orpea Group, Kaiser Permanente, and NHS entities — are cited solely on the basis of publicly documented regulatory findings, official investigations, court records, parliamentary reports, and other published primary sources. No non-public information has been used. All analysis is independent, educational, and analytical in nature.
Nothing on this site constitutes legal, regulatory, financial, investment, or professional advice of any kind. The Decision Integrity Chain™, Decision Engineering™, FUSE™, STAGE™, and related frameworks are proprietary intellectual property of Deepak Aggarwal. Unauthorised reproduction or commercial use is prohibited.
Case studies and scenarios described as "constructed" or "composite" are hypothetical illustrations based on documented failure patterns. They do not refer to any specific institution, transaction, or individual beyond what is explicitly stated.
© Deepak Aggarwal 2025–2026. All rights reserved. Decision Engineering™ · Decision Integrity Chain™ · DIC™ · FUSE™ · STAGE™ are trademarks of Deepak Aggarwal.