Diamax has been engineering information systems for institutions since 1998.
The first engagement was a multichannel publication of the World Bank's Operations Manual. The current one is a research management platform for a global pipeline program. In between: board proceedings, election systems, executive correspondence, membership records, a committee reviewing the IPCC at the request of the UN Secretary-General, and a good deal that will never be described publicly.
The problems have not changed materially. The tools available to address them have changed considerably.
Twenty-eight years. Twelve people across three countries as well as long standing specialist collaborators, most of whom have been with Diamax more than a decade. Systems in continuous production for more than twenty years. A small number of engagements each year, by design.
It is a small firm, and we prefer to state the number rather than leave it to inference.
When a problem exceeds what we know, we bring in a specialist, and it is the same small group of people we have worked with for twenty years and more, because they know how we work and we know what they are good at. They join the team; they do not take work off it. The distinction is worth being precise about. Adding a specialist does not lighten the load. It adds coordination, integration, and another set of judgments to reconcile, which makes the work better and also makes it more demanding. That is one of the reasons we take few engagements at a time, and it is the opposite of handing a project to someone a client has never met.
We never outsource.
Diamax is small because proximity, continuity, and judgment are advantages. The people a client meets are the people who lead the work, and there are no layers between the person who understood the problem and the person writing the code.
The fair question is what happens if we are not here. It is a reasonable thing to ask of any firm, and most do not answer it. Every system we steward has documented runbooks, delivered as part of the stewardship fee rather than promised for later. Source code is escrowed where a client wants it. Every system in our care has named coverage from more than one person and a documented rotation, because a commitment that depends on one individual's telephone is not a commitment.
Our senior team has been together for twenty years. That is the real answer. The rest is what we do in case it stops being true.
Founded the firm in 1998 while still working at the World Bank. The source of much of the firm's point of view.
Joined during the World Bank boardroom project and has led engineering since. Twenty years of continuity in the role, which is generally what a technical buyer wants to establish.
With the firm since 1998. Responsible for delivery and resolution across the entire arc of its work, and usually the first to know when something has gone wrong.
Behind the three of us are engineers, designers, analysts, and quality specialists in the United States, Canada, and Europe, most with the firm ten years or more. Alongside them is a small group of collaborators who have worked with us for over twenty years, brought in when a problem exceeds what we know rather than when we need additional hands.
Derek, Adrian, and Cristina are involved in every engagement, which is the reason there are only a few each year. It is a deliberate limit rather than a bottleneck: engagements are led by senior people who have run work of this kind for years, and the firm does not accept more than it can staff that way. That continuity is deliberate, and in a firm of this kind it is most of what a client is actually buying.
We would rather a prospective client spoke to someone than looked at a wall of logos.
Ask and we will introduce you to a client who has worked with us for more than fifteen years, and to one whose system we still run. They can describe what we are like when something goes wrong, which is usually the more useful question.
A short Metro ride from Washington, where most of our clients are.
Our engineering studio. It exists since 2007 because our Washington team was being recruited away and we needed continuity we could count on for decades. Costs there are comparable to Washington; this was never a cost decision. We are in Bucharest for the depth of the talent, and because talented people build careers there with one company.
The studio is open to clients. Some of the best work we have done occurred when a client team came to Bucharest for a week and worked on nothing else.
For consulting, strategy, and design, where what is being bought is judgment rather than hours.
For implementation and engineering, where scope is genuinely open and transparency serves a client better than an invented fixed figure.
Where speed matters. Development can now run at a rate that would have been impossible three years ago. How far to increase it is the client's decision, made against the consequence of the problem, and we set out what it costs and what it buys beforehand.
For the systems we continue to look after.
Analysis and design are scheduled and priced in days rather than hours, because that is the unit in which the work arrives.
We take care of what we build, for decades.
Several of these systems have outlived the staff who commissioned them, and more than once we have found ourselves training a client's new people not on our application but on their own business processes. Institutions lose their memory in ways they rarely plan for, and one of the things a long relationship provides is a place for that knowledge to remain.
A prospective client will discover this in any case, so we state it plainly.
Derek founded both companies. Milieu Gravity is a separate company with its own ownership, its own balance sheet, and its own customer contracts, and it was separated from Diamax deliberately. The separation exists so that the advice a client receives from us is not an argument for a product we are selling.
In practice, an engagement with us is a contract with Diamax. Where a client licenses the platform, that contract is with Milieu Gravity and we do not sit in the middle of it. Our custom applications are not built on Gravity; they stand alone, or integrate with it, or integrate with whatever a client already runs, and that is an architecture decision made on the merits of the problem. We explain our reasoning either way, including when the conclusion is that a client should not license the platform.
We are one of its certified implementation partners. We take the difficult implementations and recommend other partners for the remainder, particularly where they bring vertical expertise we do not have.
Where we have moved a client onto Gravity, they remain our client. Most of those relationships are a decade old or longer and we intend them to continue. The platform changed; we did not. The same holds for a new implementation: we do not hand a client over at go-live.
We develop people in the older manner: alongside someone senior, on real work, over years.
If that appeals to you, write to us. We hire rarely and carefully, and are glad to hear from people before a position exists.