Sravan Modugula
Chief Data & AI Officer, dotSolved · Creator of Nexatron
Data, AI and digital transformation leader for banking, healthcare and technology
Sravan Modugula is Chief Data & AI Officer at dotSolved and the creator of Nexatron, the agentic answer engine for governed enterprise data. A transformational leader, he leads the company’s AI and Data practice, directs Nexatron’s product and engineering work, and advises boards and executive teams on data strategy, investment and governance.
Over a 20-year career he has led technology at every scale of banking. As Executive Director at JPMorgan Chase he worked inside a global systemically important bank. Across nearly eight years at First Republic he led enterprise digital transformation as the bank grew from a community bank into a regional, super-regional and finally national institution, building the digital banking, private banking, wealth management, data and cloud platforms that each stage of growth and regulatory scrutiny demanded.
His work has included digital banking for millions of customers, real-time customer data and petabyte-scale cloud migration. At dotSolved he also works with healthcare and life sciences, genomics and diagnostics, health benefits, AI infrastructure and technology companies. His current focus includes governed Claude and MCP adoption on enterprise systems, connecting Claude to ERP and CRM data with the permissions, definitions and evidence that finance and IT teams can check. The questions he brings to these programs are practical: who owns the data, how will the result be checked, and who will operate the service?
He holds an executive certificate in Organizational Leadership from The Wharton School and is named as sole inventor on 15 US patent applications.
Banking technology leadership at every scale
Every stage of a bank’s growth brings a different regulatory bar, a different operating model and a different definition of “enterprise grade.” Sravan has led data and digital platforms through each of them.
Community banks
Modernizing core and digital banking channels, building a first enterprise data foundation and putting in place the controls examiners expect, at a pace and budget a community bank can sustain.
Regional and mid-market banks
Scaling platforms, data governance and risk management through rapid growth and each new tier of supervision: regulatory reporting, BSA/AML data, stress testing readiness, model risk management and third-party risk.
National and global banks
Digital banking and wealth management platforms for millions of customers, high-availability customer data, petabyte-scale cloud data platforms and enterprise AI governance at the standard of a global systemically important bank.
Leadership record in regulated industries
Regulated industries
Commercial banking, private banking and wealth management under OCC, Federal Reserve, FDIC and state supervision. Healthcare, health benefits, genomics and life sciences under HIPAA and patient privacy obligations. Compliance, audit evidence and model risk designed in from the first architecture review.
Transformations led
Enterprise digital transformation, cloud migration and data warehouse modernization, real-time customer data and master data management, data governance programs and enterprise AI adoption, each delivered without disrupting the systems of record a business depends on.
Board and C-suite partnership
AI and data strategy for boards and executive committees, technology risk posture for audit and risk committees, and investment, operating model and build-or-buy decisions with CEOs, CFOs, CIOs, CTOs and CISOs.
Product creation and invention
Creator of Nexatron, the agentic answer engine for governed enterprise data. Named as sole inventor on 15 US patent applications covering trusted AI answers, governed data access and consistent business meaning.
Signature work in banking data and digital platforms
Always-on customer data for digital banking
Online and mobile banking depended on a master data management platform that reconciled overnight, so an intraday problem could leave customers unable to see their own accounts. Sravan led an event-driven architecture that kept nightly reconciliation as the trusted baseline and streamed intraday changes to a real-time customer data store. Customers saw current information, and none of the enterprise platforms underneath had to be rebuilt.
An enterprise data integration platform ahead of the market
He built an internal platform for reliable, governed data pipelines between operational systems and analytics years before managed data integration products became mainstream, giving the bank faster access to trustworthy data at a fraction of the effort.
Petabyte-scale cloud data modernization
He led the migration of petabyte-scale data estates from on-premises data centers to a cloud data platform and modernized the enterprise data warehouse, preserving lineage, controls and reporting continuity throughout.
Digital banking and wealth platforms for millions of users
As the bank grew, he led the digital banking, private banking and wealth management platforms behind client and advisor experiences, scaling performance, security and regulatory compliance in step with the business.
From enterprise platforms to an AI product
At dotSolved he turned two decades of lessons about business definitions, permissions and evidence into Nexatron, so people can ask questions in plain English and receive answers that respect the governance already in place.
Leadership approach to enterprise AI programs
Sravan is a transformational leader. He works with business and technology leaders to agree what a program needs to accomplish, who can make decisions and how the team will judge progress. He develops the leaders responsible for delivery and for running the service after launch.
Solve the organizational problem first. The companies that move fastest settle ownership, decision rights and data governance before choosing technology. Most stalled programs are alignment problems presented as technical ones. The enterprise AI governance framework and template sets out the ownership, decision rights and controls to settle first.
Define the decision before choosing the model. A revenue question needs an agreed period, population and calculation. Model quality cannot settle a disagreement about what the business means by revenue. Nexatron, the governed data intelligence platform, applies this rule by attaching the query and the certified metric definition to every answer.
Measure the complete workflow. Include preparation, review and exceptions when judging an AI program. Faster generation is useful only when the completed work meets the business requirement.
Five stages of enterprise AI maturity
Sravan uses this framework with executive teams to distinguish successful experiments from services the organization can support and expand. The move from departmental wins to a shared platform calls for explicit ownership, agreed definitions and a working support model. Executive teams that want to locate their organization on this scale can request an AI readiness assessment.
Stage 1
Experimentation
Scattered pilots, different tools and no committed budget.
Stage 2
Departmental wins
One team ships something valuable, but without shared governance.
Stage 3
Platform foundation
Shared data platform, central team and working governance.
Stage 4
Operationalized AI
AI embedded in core processes and tied to business results.
Stage 5
AI-native organization
AI informs strategy, pricing, risk and customer experience.
US patent applications arising from Nexatron
Sravan is named as sole inventor on 15 US patent applications arising from Nexatron. They address a single question: how can a business trust an answer produced by AI from its own data? Areas include:
- Calibrated confidence for AI-generated answers
- Self-correcting business metric definitions
- Plain-English data governance policies
- Version control and impact analysis for metric definitions
- Consistent metrics across multiple data warehouses
- Adversarial testing of natural-language analytics
- Privacy-preserving learning across deployments
- Causal explanation of metric changes
- Transparency reporting for AI models
- Organizational knowledge from analytical questions
Selected writing on enterprise AI decisions
Six starting points for executives and engineering leaders evaluating enterprise AI.
Executive decisions
What a CIO Should Ask Before Connecting Claude to Business Data
The questions a CIO should settle before giving an assistant access to business information.
Integration design
Building an MCP Connection for a Business Application That Has None
How to scope a useful MCP adapter when an application has no native connector.
Model evaluation
Anthropic or OpenAI? Evaluate the Enterprise Workflow Before the Model
A workflow-based comparison of providers, with evidence that a business owner can review.
Infrastructure decisions
NVIDIA GPU Planning Starts with Accepted Tasks, Not Peak Throughput
Evaluate GPU investment against accepted tasks, operating costs and the intended workload.
Operating results
Measure Adoption Without Rewarding Empty Activity
Separate activity statistics from improvements in the work people need to complete.
Product evaluation
Prepare Three Business Questions for a Nexatron Demonstration
Use a familiar question, an ambiguous question and an access boundary to evaluate Nexatron.
Areas of expertise
- Enterprise data and AI strategy
- Digital transformation for community, regional and national banks
- Private banking and wealth management technology
- AI and data governance for banking and healthcare
- Cloud data platform modernization and migration
- Master data management and real-time customer data
- Enterprise Claude and MCP adoption
- AI product creation, model evaluation and MLOps
Education
The Wharton School, University of Pennsylvania · Executive certificate, Organizational Leadership