I have spent more than 25 years helping organizations solve difficult business, technology, regulatory and transformation problems.

Increasingly, one question has connected much of that work:

How do we use technology to create better economic outcomes without continuing to increase the resources required to achieve them?

That question ultimately led me to co-found Exemplary Solutions Group — ESG-i.

We established ESG-i in Canada in 2024 with an ambitious but practical objective.

Environmental, Social and Governance programs were generating strategies, assessments, commitments and reporting requirements, but organizations still struggled with the operational machinery required to turn those commitments into measurable outcomes.

We saw an important gap between ESG strategy and ESG execution.

Our original vision was therefore not simply to create another ESG consultancy.

We began building an ESG Investment and Consulting Service Bureau capable of helping organizations understand their environmental position, determine where investment could create measurable improvement, collect the evidence required to support those decisions, manage sustainability initiatives and ultimately support operational and regulatory reporting.

My background in financial services, government, regulatory systems, enterprise architecture, quality management and large-scale technology transformation heavily influenced that approach.

I have always believed that governance and reporting are strongest when they are connected directly to the systems, processes, assets and evidence that create the underlying business outcome.

That meant ESG-i needed technology.

Not another dashboard.

Not another sustainability spreadsheet.

We needed an operational technology backbone.

That requirement led us into what became a much deeper collaboration with MEISTRO AI.

Our early work concentrated heavily on buildings, facilities and infrastructure because these environments make the ESG challenge tangible.

A building contains electrical systems, HVAC equipment, automation controllers, sensors, engineering drawings, operating schedules, maintenance histories, energy consumption and thousands of other interconnected pieces of information.

Yet much of that information traditionally resides in different systems, documents and organizations.

We began applying AI, automation, private data management and structured knowledge to connect those pieces.

MEISTRO developed into the technology architecture that allowed us to work with engineering information, building automation data, assets, energy information, operating conditions, documentation, workflows and expert knowledge as parts of one contextual environment.

Instead of simply asking an AI model a question, we became increasingly focused on giving AI the correct business context in which to reason.

That distinction became fundamental to my work.

Today, MEISTRO combines private enterprise information, domain knowledge, structured context, workflows, Playbooks, automation and multiple forms of AI into a governed enterprise environment.

My role as a Co-Founder and Chief Design Architect has concentrated on the business and solution architecture behind that environment: determining how professionals will actually use AI, how context will be preserved, how applications will work together and how AI can augment experts without turning critical business processes into uncontrolled experiments.

There was an irony in what we were building.

We were using AI to help organizations reduce waste, optimize buildings, improve energy performance and make better sustainability decisions.

At the same time, the global expansion of AI was creating a rapidly growing demand for computing infrastructure, electricity, cooling, memory and data-centre capacity.

That changed the ESG problem for us.

We could no longer look only at how AI could help organizations become more sustainable.

We also had to ask:

How do we make AI itself more sustainable?

That question has become one of the most important areas of my work.

Our work at ESG-i has now evolved into the development of ESG-i ROSE, a new approach to improving the efficiency of AI training and computational execution.

ROSE addresses the problem below the application layer.

Instead of assuming that increasingly capable AI must always require proportionally more computation, memory, electricity and infrastructure, we are examining how the mathematical and computational work itself can be performed differently.

The objective is straightforward:

Achieve more useful AI work from the computing resources we already have.

That has important implications.

Reducing unnecessary computation can mean:

• lower electricity demand;
• lower cooling requirements;
• reduced pressure on GPU and memory resources;
• lower infrastructure requirements;
• lower operating costs;
• faster experimentation; and
• the ability to perform more research within the same compute envelope.

The environmental opportunity and the research opportunity are therefore directly connected.

I am particularly interested in what this can mean for researchers, universities, AI companies, engineers and other specialists developing the next generation of artificial intelligence.

More efficient computation does not only reduce energy.

It can increase the amount of research that can be performed with a fixed amount of infrastructure.

Researchers can test more hypotheses.

Engineering teams can run more experiments.

Smaller organizations can investigate problems that might otherwise require infrastructure beyond their reach.

Experts can spend more of their resources advancing the science rather than simply purchasing more computing capacity.

That democratization of computational capability could ultimately become as important as the energy savings.

We started ESG-i because we wanted to help organizations make better sustainability investments and convert ESG commitments into measurable operational outcomes.

To deliver that service effectively, we needed a sophisticated technology platform.

That requirement helped drive our work with MEISTRO AI.

Building MEISTRO forced us to understand AI architecture, enterprise context, private information, automation, knowledge management and computational behaviour at a much deeper level.

That work eventually exposed another sustainability problem:

the resource requirements of AI itself.

Today, we are applying what we learned to that problem.

For me, that is the natural evolution of ESG-i.

Sustainability cannot simply mean measuring environmental impact after it occurs. We must continue finding better ways to perform the underlying work.

That applies to buildings.

It applies to businesses.

And increasingly, it applies to artificial intelligence.