I build connected platforms that turn fragmented identity, data, integration, and developer capabilities into reusable enterprise ecosystems.

The problem

California needed a scalable, compliant way to manage patient consent across a fragmented behavioral health ecosystem. Consent could not function as a standalone feature — it depended on trusted identity, people matching, secure access controls, and interoperable workflows across EHRs, partner organizations, APIs, and digital channels. Without a shared platform model, the state faced risks around inconsistent consent enforcement, duplicate records, weak auditability, and poor interoperability.

My approach

I led product strategy across the Consent Management Platform, enterprise identity and access management, and related client management capabilities. I shaped the roadmap for consent capture, retrieval, revocation, and auditability across web, mobile, API, and EHR-integrated workflows. In parallel, I defined foundational direction for canonical models, master client profile strategy, people matching, and identity/access patterns including RBAC, ABAC, SSO/OIDC, MFA, and delegated administration — aligning engineering, architecture, policy, legal, and operations on source-of-truth decisions, service boundaries, FHIR integration, and secure onboarding. I also used Generative AI tools to accelerate requirements, feature design, and technical flow prototyping.

The outcome

A stronger statewide foundation for patient consent management, built as a governed platform capability rather than a point solution. The work established a clearer path for secure patient access, consent enforcement, canonical identity, and interoperable data exchange across a highly regulated ecosystem — laying durable foundations to support future scaling across patient, provider, payer, and platform workflows.

The problem

Arvest needed to modernize customer identity and access across retail banking, treasury management, and support channels. Legacy authentication and enrollment patterns created friction for customers, operational complexity for the bank, and limitations in how fraud controls, MFA, and secure access policies could scale — all without breaking existing customer experiences or downstream operational support.

My approach

I led product strategy across CIAM modernization, including authentication, enrollment, MFA, fraud-aware access, and customer identity workflows. Partnering with engineering, security, operations, and business teams, I defined the roadmap for modernizing identity capabilities and improving secure access patterns. This included driving migration to Ping Identity and Ping Protect, refining authentication policy and risk-aware controls, and aligning identity workflows with support and service operations — balancing customer experience, security posture, regulatory expectations, and platform consistency while reducing migration risk.

The outcome

A more scalable and secure identity platform strategy that strengthened authentication, policy-based access, and fraud-aware customer journeys across digital banking. The work improved the foundation for MFA, enrollment, and secure access while creating more consistent identity patterns across channels in a regulated financial services environment.

The problem

Engineering teams across the enterprise needed better ways to discover, access, govern, and reuse shared data and platform services. Without stronger platform capabilities, teams duplicated effort, onboarded inconsistently, and slowed delivery through fragmented patterns for shared services. Capital One needed a more scalable developer and data platform approach that balanced usability, governance, secure access, and time-to-market.

My approach

I led product strategy for enterprise developer and data platform capabilities supporting service discovery, metadata visibility, lineage, governed access, and reusable service adoption. My work shaped developer portal capabilities, lifecycle patterns for shared services, CI/CD-adjacent onboarding and delivery models, and platform governance frameworks that improved how teams discovered and consumed APIs and data products. I partnered across product, engineering, architecture, and platform stakeholders to reduce fragmentation, improve onboarding, and increase adoption of reusable services at scale.

The outcome

The platform scaled to support more than 200 internal and external services and 20,000 active users, contributing to 20% platform growth over two years and reducing time to market by six months. More broadly, the work shifted the organization toward reusable, governed platform patterns rather than one-off implementations — improving developer productivity, platform adoption, and enterprise leverage.

The problem

QuickBooks customers often needed help during high-intent, time-sensitive moments — tax preparation, account management, product troubleshooting. Traditional help content and static support experiences were not surfacing the most relevant guidance fast enough, which increased customer effort, reduced self-service success, and drove unnecessary assisted support costs. The challenge was to improve digital support in a way that felt more personalized, intelligent, and scalable across a global customer base.

My approach

I owned the global self-service and digital help platform for QuickBooks, leading product strategy for AI-powered support, FAQ, and personalized recommendation experiences. My work used behavioral signals, customer context, and intent to improve how support content was surfaced and consumed. I guided the integration of generative AI, semantic retrieval, and recommendation-based capabilities to create more relevant, conversational help — partnering across engineering, data science, design, and content teams to define workflows, quality guardrails, and success metrics, ensuring solutions improved real customer outcomes rather than adding AI for novelty.

The outcome

A more intelligent, personalized self-service experience that increased targeted impressions by 8% and generated $1.9M in month-over-month revenue. The work drove a 25% increase in site traffic, improved end-of-year tax activity by 8% across product suites, reduced contact rate by 3%, and contributed to a 15% gain in operational efficiency — moving QuickBooks support from a static help model toward an adaptive, AI-assisted experience designed around the moments that matter.

The problem

Youth sports registration is highly fragmented across municipal recreation departments, school athletics sites, independent leagues, social media, PDFs, email lists, and word of mouth. Parents often don’t know where to look, when registration opens, or which organizations serve their child’s age or grade — creating unnecessary friction that can disadvantage families newer to a community or less connected to existing sports networks.

My approach

I designed Open Play Sports as a centralized discovery and registration-intelligence platform that helps parents find youth sports opportunities based on location, age, grade, season, and activity. Rather than building another league-management or registration system, I positioned Open Play as a vendor-neutral discovery layer that aggregates programs across municipal recreation systems, schools, clubs, and independent sports organizations. I defined the product strategy, MVP requirements, canonical activity data model, search and eligibility logic, notification model, and source-provenance framework — and designed an AI-assisted ingestion architecture that collects structured and unstructured information from sources such as MyRec, SportsEngine, WebTrac, CivicRec, and municipal websites, normalizes it into a common schema, and routes uncertain data through human review. The product is being built using an AI-assisted development workflow with a web-first architecture designed to support future mobile applications and shared platform services.

The outcome

Open Play Sports has progressed from problem discovery to a functioning web product with activity search, eligibility filtering, program detail pages, registration links, alerts, organization discovery, community submissions, and data-verification concepts. I established an initial Vermont source registry spanning municipal recreation departments, statewide sports associations, and independent youth organizations to support automated program discovery and future statewide coverage. The next phase is a Central Vermont pilot designed to validate program coverage, registration click-throughs, alert adoption, information accuracy, and whether families discover opportunities they otherwise would have missed.