Selected work

Software with
a water-domain foundation.

Most of my work lives inside a drinking-water utility. These summaries describe my contribution and the stage of the work, with public references where available.

01 / Digital water

Making water operations easier to reason about

Concepts & working prototypes

Digital-twin concepts that connect operational data with scenarios, replay, and validation.

The question

Drinking-water decisions draw on measurements, models, and the experience of domain experts. Bringing those together requires a shared understanding of what the data means and where its limits are.

My contribution

I work with domain experts on digital-twin concepts for drinking-water treatment and distribution. I build workflows for inspecting data, comparing scenarios, and checking assumptions, alongside reusable tools for analysis.

Where it stands

This work produces prototypes and shared patterns that help domain experts and data teams evaluate what is useful. Validation and operational ownership remain part of deciding what can move into production.

A model is useful when the people making decisions can check its inputs, assumptions, and results.

02 / Internal software

A shared starting point for data work

In use across the data team

Internal Python SDKs that give a growing data team reusable access methods and domain logic.

The question

Repeated data-access code makes analysis harder to maintain. Domain experts need understandable interfaces to governed data, with examples they can use in their own work.

My contribution

I build and maintain internal Python SDKs with standardized access methods, reusable domain logic, documentation, and examples. I also help colleagues contribute through code review and coaching.

Where it stands

The tooling has become a shared foundation across a growing data team, supporting analysis, automation, and internal applications. Improvements can be reused across projects.

The next project should benefit from what the team has already learned.

03 / Applied AI

Helping a utility put AI into practice

Exploration, guidance & adoption support

Experiments, practical guidance, and rollout support with human review and confidentiality in mind.

The question

Using generative AI in a utility raises practical questions about confidentiality, reliability, and responsibility. Teams need ways to explore it and judge where it helps.

My contribution

I ran a multidisciplinary exploration group, translated use cases for management and technical colleagues, and contributed to responsible-use guidance and ChatGPT and Copilot pilots.

Where it stands

That work supported a shared understanding of useful applications and their limits, including a Copilot pilot that developed into an organization-wide rollout.

People need to know what they are responsible for checking before they rely on an AI-assisted workflow.

Further reading

Public references

A roadmap to digital twins for drinking-water treatment

KWR / BTO · 2023

I contributed domain and implementation input and am acknowledged as part of the report's steering committee.

Generative AI in water utilities

KWR · 2024

Sector context for practical AI use, including extracting information from operational plans.

Mobile water monitoring in Thailand

TU Delft · Fieldwork

A student field project on mobile water monitoring in the Phetchaburi river basin.

Contact

Working on a related problem?

I'd like to hear about it. Get in touch about water infrastructure, internal software, or putting AI to practical use.

Email meContact details