Case Studies

Much of my strongest work is internal, so I describe the transferable capability rather than sensitive implementation details. The public signal is in generalized case studies, sector references, and architecture patterns: digital twins, reusable data foundations, governed AI adoption, and software that helps infrastructure teams reason about complex physical systems.

Selected Work

Building Digital-Twin Capability For Water Operations

Moving from dashboard thinking toward inspectable digital-twin capability for drinking-water treatment and distribution, while keeping validation, governance, simulation readiness, and production ownership visible.

Context

Water operations contain a lot of tacit knowledge: treatment assumptions, distribution topology, laboratory context, operational telemetry, source metadata, and local reasoning that rarely lives in one coherent system.

Build

I translated that context into web-based workflows for inspection, replay, scenarios, reporting, topology, metadata, and data-quality reasoning. The implementation pattern combines React/TypeScript interfaces, Python/FastAPI services, relational data, cloud services, reusable internal SDK layers, validation logic, and controlled paths toward domain-solver integration.

Result

The work created a serious platform nucleus rather than a static dashboard. It made the future capability concrete enough for domain experts, data teams, architecture stakeholders, and leadership to discuss what should be hardened next.

What matters next

The next phase is not only technical. A digital twin becomes risky if it is promoted faster than the organization can own, validate, deploy, and maintain it. I deliberately treat production readiness, internal SDLC, documentation, access control, testing, scenario lineage, and operational ownership as part of the product.

Internal Python SDKs As Data Infrastructure

Creating reusable Python foundations that let domain experts and data teams work with enterprise data without rebuilding the same access logic in every notebook, script, or app.

Context

Enterprise data platforms are powerful, but the distance between governed data architecture and day-to-day analytical work can remain too large. Domain experts need stable, understandable interfaces, not just access to raw tables or isolated dashboards.

Build

I built internal SDK patterns that sit above enterprise data-platform layers and APIs: standardized access methods, reusable domain logic, documentation, examples, and application-friendly interfaces.

Result

The SDK work reduces fragile one-off scripts, accelerates analysis, and gives internal software a more reliable foundation. It also creates a practical bridge between data-platform teams and operational users.

What matters next

The long-term value is cultural as much as technical: shared interfaces make stronger engineering practices easier to adopt because reuse, review, documentation, and typed contracts become part of normal data work.

Governed GenAI Adoption

Helping move generative AI from curiosity toward governed practical use through early LLM experimentation, a multidisciplinary focus group, briefings, responsible-use input, Copilot workflows, and AI-assisted engineering.

Context

Generative AI created immediate opportunity and immediate governance questions. In a utility context, usefulness is not enough: risk, confidentiality, quality, explainability, and adoption patterns matter.

Build

I identified LLMs early as a transformative technology, initiated a multidisciplinary focus group to broaden practical experimentation, helped shape internal exploration around ChatGPT and Microsoft Copilot, translated use cases for management and technical audiences, and developed practical ways to use AI as engineering leverage while keeping human judgment explicit.

Result

The work helped build organizational confidence around where GenAI is useful, where it is risky, and how it can fit into internal software, documentation, analysis, and knowledge workflows.

What matters next

The important distinction is between black-box automation and inspectable augmentation. I prefer AI-assisted workflows where the reasoning, code, data assumptions, and validation steps remain visible.

Public References & Sector Work

KWR / BTO Digital Twin Roadmap

KWR, 2023

Public sector context for digital-twin thinking in drinking-water treatment. I contributed substantial domain, product, and implementation input and am acknowledged as part of the steering committee.

Operating Principles

  • Start with the operational decision, not the dashboard.
  • Use AI to increase throughput without hiding responsibility.
  • Prefer inspectable software models over impressive but opaque automation.
  • Treat governance, documentation, and ownership as part of the product.
Read the full approach

What I Am Interested In Next

I am interested in applied AI and data-platform work where the stakes are practical: water, infrastructure, climate adaptation, operational intelligence, and tools that help expert teams make better decisions. I am especially interested in wrapping domain solvers such as PHREEQC and EPANET, and integrating adjacent domain engines where useful, in software workflows that make inputs, assumptions, validation, and results easier to inspect.

Start a conversation