Skip to main content
Enterprise AI, decodedJuly 2026
GroundingNodes

About GroundingNodes

I operate at the intersection of enterprise strategy and practical AI engineering — the critical, often messy phase where a model that shines in a demo is hardened into a system the business can rely on day to day.

My approach is grounded in more than 19 years of experience across the technology lifecycle. Over the course of my career, I have worked across software engineering, delivery management, technical pre-sales, and enterprise account leadership in Finance, Automotive, and Retail. That breadth has given me a unique perspective on how complex technology is evaluated, sold, implemented, and sustained at scale. What I have learned is that enterprise AI programs rarely fail because of the underlying models. They fail on grounding, trust, and adoption:

  • The Trust Trap: A dashboard that displays three different numbers for the same metric to three different leaders loses organizational trust in a single meeting — and a stronger LLM won't win it back.
  • The Adoption Gap: A tool ships with full backing from the top but sits completely unused because the nuance of the end-user's workflow was left out of the engineering requirements.

    • Across the AI programs I have helped bring into production - from voice agents and warranty automation to lead optimization, appointment recovery, and analytics dashboards - the questions that ultimately determine success are rarely about the model itself. The questions that matter to adoption are entirely operational:

  • How do we force a probabilistic LLM to respect deterministic business rules?
  • How do we capture the real requirements for building the agent?
  • Does this architecture actually move the contribution margin per transaction, or is it just compute-heavy novelty?

    • My philosophy is simple, if it doesn't move the math, it shouldn't be built.

    An agent is only as effective as the systems it can interact with and the data it can access. Organizations operating within fragmented technology estates or vendor ecosystems that limit interoperability by not exposing APIs often discover that integration, and not intelligence, becomes the primary bottleneck.

    This is why decisions about orchestration, tooling, infrastructure, data governance, and API strategy matter so much. They determine whether AI remains a collection of isolated experiments or evolves into a scalable enterprise capability. The strategic choices made today will shape operational flexibility, cost structures, and competitive advantage for years to come.

    Write to the author

    § 01

    The Bridge

    Projects stall when business teams speak exclusively in outcomes and engineering teams speak exclusively in architecture. Because they talk past each other, promising solutions get trapped in prototype purgatory. With an engineering foundation and an MBA from the Indian School of Business, I am equally comfortable discussing contribution margins and operating models with executives as I am debating architectures, evaluation frameworks, and implementation trade-offs with engineering teams. Success requires both perspectives.

    To deepen that expertise, I made a conscious decision to stay close to the technology itself rather than observe it from a distance. I publish the research I conduct in my free time on this blog - GroundingNodes, to document these experiments and decode complex architectural signals for my network. In it, I strip away the vendor pitch from AI prouducts to deliver the honest, data-backed blueprints required to take AI out of the sandbox and build it into the permanent operating fabric of businesses. I also regularly review technical research, architecture papers, and implementation patterns to identify the approaches most likely to scale in real enterprise environments.

    § 02

    Beyond the Enterprise

    Based in Rotterdam alongside my husband, I balance my professional life with creative and grounding pursuits. In my free time, I paint acrylic landscapes, enjoy long walks, love meeting new people, and read widely to keep pace with where the technology and the market are heading.