Where Data & AI Strategy, Data Governance and AI Governance come together to move AI from experimentation to responsible, scalable execution.
Before the Prompt is a research and community platform built for Data and AI leaders and practitioners.
We explore the foundational challenges behind enterprise AI — trusted data, AI Governance, processes, adoption and responsible execution — through research, practical frameworks and small peer-level conversations.
Built from 13+ years inside Goldman Sachs, Morgan Stanley, UBS, and Marsh McLennan, and a 1,000+ member executive community that won't settle for consulting-speak or recycled frameworks.
Practical thinking on AI Governance, Data Governance, prompt systems, and the foundations required to scale trusted AI.
Follow on LinkedIn →Deep conversations with practitioners solving real AI-readiness problems at enterprise scale. No guests paid to be polite.
Coming SoonInvite-only sessions where CDOs and AI startup founders go deep on data foundations. Three seats. Sixty minutes. Zero decks.
Explore →A curated weekly view of the enterprise AI developments that are actually changing how organizations operate — connecting adoption, governance, model economics, data foundations, agents and risk into the larger signals leaders should watch.
Explore Enterprise AI Signals →Led enterprise data governance and AI strategy across the organization, building frameworks for semantic governance, data quality measurement, and AI readiness assessment. Architected the foundational thinking that became the BRIDG·E framework.
Led data quality programs and credit risk data management, ensuring the accuracy and reliability of risk-critical data across trading and lending operations. Built quality measurement disciplines that tied remediation directly to business outcomes.
Managed product data across the enterprise and led the data incident review process — identifying root causes, driving remediation ownership, and building the operational muscle for continuous data quality improvement.
Delivered regulatory reporting and compliance data programs within one of the world's most demanding financial data ecosystems, where data accuracy carried direct regulatory and reputational consequence.
Where can AI be embedded into existing enterprise processes to improve how work gets done?
Exploring Prompt Governance, Prompt Optimization and Prompt Evaluation — including how we measure and continuously improve prompts.
Translating AI regulation, governance and emerging requirements into practical enterprise controls, accountability and documentation.
Practical frameworks and research exploring the challenges organizations face in building trusted and scalable AI.
Five pillars. Three layers. The enterprise AI readiness architecture spanning Build Governance, Remediate Quality, Institutionalize AI Governance, Drive Change, and Grow AQ + Empower EQ — built from 12+ years inside Wall Street and global insurance.
Explore the Framework →The organizational limbo where AI pilots succeed in demos but never reach production — because nobody fixed the data that sits beneath them. Named. Analyzed. Mapped to root causes.
Explore on LinkedIn →Seven pillars for measuring and developing the human capacity to work alongside AI — Curiosity, Critical Thinking, Unlearn-to-Relearn, Data Ethics, Collaborative Intelligence, Decision Agility, Data Storytelling. Beyond data literacy.
Explore on LinkedIn →A 4-stage semantic resolution methodology: Collect → Extract → Cluster → Resolve. The structured antidote to AI systems that don't know what "revenue" means — because your enterprise doesn't either.
Explore on LinkedIn →The hidden accumulation of undocumented, manual, and brittle processes that silently collapse when AI workloads arrive — before anyone notices. The governance gap nobody tracks until it's too late.
Explore on LinkedIn →A curated weekly view of the enterprise AI developments that are actually changing how organizations operate — connecting adoption, governance, model economics, data foundations, agents and risk into the larger signals leaders should watch.
Enterprises are increasingly matching models to workloads based on capability, cost, latency and risk instead of standardizing on a single frontier model.
Token usage, inference cost, routing, caching and cost-per-task are becoming management questions — not just technical implementation details.
As AI systems begin taking actions, governance is shifting toward identity, permissions, tool access, monitoring, audit trails and escalation at execution time.
The full Enterprise AI Signals briefing brings together the week’s strongest public developments, explains the trends connecting them, and highlights the terminology, implications and questions enterprise leaders should be watching.
Senior data leaders and startup founders. One topic. Sixty minutes. No slides. A small, curated series built around the real problems behind AI readiness — how people are actually solving them, what's working, and what's not.
"The conversations that change how you think about a problem don't happen in conference halls. They happen at small tables."Request a Seat →
Nikunj Desai is a Data & AI Strategy leader with 13+ years of experience across Goldman Sachs, Morgan Stanley, UBS and Marsh McLennan. His current work focuses on trusted data, AI Governance, AI-enabled processes and the foundations required to scale AI responsibly.