Prompting Scales. Verifying Doesn't. Yet.
I've been thinking about something Balaji Srinivasan wrote in June: prompting scales, because prompting is just typing, but verifying doesn't scale, because verifying requires much more than typing.
"I am grateful to Oracle for the nomination, and humbled by the opportunity to be considered by the W3C community for the Board of Directors. For more than three decades, W3C standards have helped shape the Web. Today, as AI and new computing architectures reshape the digital ecosystem, our challenge has expanded: we must strengthen the institution that keeps the Web open, interoperable, accessible, and safe. Delivering on that Vision requires more than technical excellence. It requires disciplined priorities, transparent governance, financial resilience, broad stakeholder support, and the ability to anticipate change without losing sight of the mission. I bring over 16 years of experience building and leading technology across enterprise software, cloud, AI, accessibility, and the Web. At Intertrust, Autodesk, Amazon Web Services, and Oracle, I have worked where strategy must translate into outcomes, risks must be surfaced early, priorities compete for resources, and decisions must stand up to real-world scale. I also bring a global perspective. Born and raised in India, and having lived and worked in the United States for more than 17 years, I have led and worked with teams across the United States, Canada, the United Kingdom, Poland, Estonia, Ireland, Mexico, and India. This experience has given me a practical understanding of how differently access, participation, and opportunity are experienced across regions. A worldwide Web requires an institution capable of recognizing those differences while building durable common ground. My work with W3C's Accessibility Guidelines Working Group provides another vital lens. I understand the rigor of consensus-based standards development, while much of my career has been spent on the implementation side, turning standards and principles into products and platforms that work at enterprise scale. My book, Principles and Practices of Front-End Engineering, reflects that same focus on translating sound engineering principles into practice. W3C's 2025-2028 Strategic Objectives set the right direction. The test now is disciplined stewardship: translating those objectives into clear priorities, measurable outcomes, appropriate resources, and accountability. My experience as a builder and implementer gives me a practical perspective on three priorities for the Board. First: Broaden Participation, Strengthen Independence. W3C's institutional independence is strengthened by the breadth and commitment of the community that sustains it. The Board should ensure that W3C has a deliberate stakeholder strategy, with clear priorities and meaningful measures of progress. We should identify the communities W3C needs to engage more deeply, understand what limits participation, and strengthen the value proposition for sustained involvement. Enterprise developers, cloud and AI builders, startups, public-sector organizations, and industries increasingly dependent on the Web all have a stake in W3C's future. A broader stakeholder base brings expertise, legitimacy, resources, and resilience. I will advocate for ensuring that W3C's strategic priorities are reflected in its resource decisions and management goals, with meaningful measures of stakeholder engagement and retention. Second: Measure Impact, Not Activity. W3C has rightly made impact a strategic priority. The Board should look beyond what W3C produces to understand what changes because of its work: adoption, interoperability, accessibility, implementation, stakeholder reach, and real-world influence. These are questions of organizational effectiveness, not technical direction. I will advocate for disciplined oversight and meaningful measures that show whether W3C's priorities and resources are advancing its mission. Good governance distinguishes activity from impact. My experience on both sides of this equation, contributing to standards and implementing technology at scale, gives me a practical perspective on that distinction. Third: Govern for Long-Term Resilience. Long-term stewardship requires distinguishing temporary technology cycles from structural shifts that could affect W3C's mission, relevance, independence, or sustainability. AI is one such shift today, but it will not be the last. My experience across enterprise AI, cloud, and large-scale technology platforms gives me a practical lens for assessing how these shifts could affect W3C's relevance, stakeholder base, and long-term resilience. The Board should ensure that consequential changes are recognized early, important risks and dependencies are surfaced, and W3C's strategy remains responsive as the external environment and stakeholder expectations change. The long-term question I would keep before the Board is: What do these shifts mean for W3C's ability to remain independent, relevant, resilient, and influential ten years from now? Accessibility is one of the clearest tests of that stewardship. Institutional decisions can shape who participates and who benefits. W3C's principle that the Web is for all humanity is therefore more than an aspiration to me. It is a standard against which our decisions should be tested. Good governance begins with clarity of roles. The Board provides oversight, judgment, direction, and support, while management is responsible for execution. A strong Board does not need to be operational to be consequential. Its value lies in safeguarding resources, holding the organization accountable, surfacing risks and opportunities early, and exercising judgment about when scrutiny, intervention, or restraint is required. If elected, my fiduciary responsibility would be to W3C. I would exercise that responsibility with care, loyalty to W3C's interests, and respect for its mission, bylaws, and legal obligations. My decisions would be guided by what best serves W3C's long-term health, not the interests of any single company, technology, geography, or constituency. The Board already possesses deep institutional and governance expertise. I offer a complementary perspective: an enterprise-scale builder and implementer, a global accessibility advocate, an active W3C contributor, and a technology leader accustomed to translating strategy into measurable outcomes across complex organizations and cultures. W3C has set an ambitious direction. The responsibility now is to keep that strategy actionable, measure its impact, broaden the community that sustains it, surface risks early, and preserve the institution's strength and independence for the long term. I would be honored to earn your support."
Read & listen to full statement
I am grateful to Oracle for the nomination,...

A technically rigorous guide covering semantic HTML5, modern JavaScript, browser rendering, accessibility (ARIA and WCAG), performance, testing, and component architecture. Bridging academic learning with real-world production work.
Get it on Amazon

Shows how modern LLMs like Grok, GPT-5, and Claude can guide you through implementing accessible patterns step by step, from landmarks and labels to keyboard interactions and error handling. Practical prompt recipes, real code examples, and simple guardrails that keep automated tools and human review in the loop. Whether you are a designer or developer, you will leave with a workflow that makes WCAG compliance part of everyday practice.
View session
AI is changing front-end work, but what matters to product teams is where it actually improves outcomes. Personal case studies from building on large web platforms covering real decisions, tradeoffs, and results. Covers where AI helped with interface design, accessibility reviews, component scaffolding, and UI telemetry, and where it introduced friction or false confidence.
Watch recording
Explores how AI is rapidly automating repetitive tasks across industries, including front-end engineering. Breaks down how AI agents are reshaping workflows, with a practical focus on the real transformations happening in project management, product teams, and engineering practice.
Watch recordingActive contributor to web standards, ensuring the future of the platform is open and accessible.
Deep expertise in implementing and auditing for strict accessibility compliance requirements.
Over a decade of shipping production-ready, performant frontend architectures at scale.
I've been thinking about something Balaji Srinivasan wrote in June: prompting scales, because prompting is just typing, but verifying doesn't scale, because verifying requires much more than typing.
I didn't plan to write a book. There was no grand announcement, no sabbatical, no clean block of time carved out for writing. Just a quiet morning after my second daughter, Anjali, was born.
I've been thinking a lot lately about how AI is changing engineering teams. Not in theory, but in the actual day-to-day work.
I am currently available for consulting engagements, accessibility audits, and team training workshops.