How it works

Dromologue sells software, advice on where to use it, and the engineers who install it. The software has four parts: discipline skills that guide each team as it builds, agents that take the steps needing judgement, a digital twin of the operating model that every release updates, and a simulator that runs the twin forward. It is installed one team at a time, and none of the data it records leaves your firm.

The operating model AI needs cannot be drawn in advance, because what agents can do changes every quarter. So the software works release by release. Before a release, the sponsor sets a target on each of four metrics, and the skills make the change inside the team’s own build. Afterwards the twin shows whether the release moved the targets, and the next targets are set from what it shows.

The advice that comes first

Four pieces of advisory work come before any team installs the skills. The first is AI strategy. With the sponsor we state what the firm is for and the one problem its strategy has to solve, map its capabilities in its own words, and judge which of them set the firm apart now and in three years and which do not. We draft the choices that follow; the sponsor makes each one, and each carries an objective whose results can be read within the first increment. The second is the practices review, in which the sponsor and the nine discipline owners take every principle and practice of the nine disciplines in turn and adopt, amend or reject it, adding their own where the firm needs them. The set that results is the one the firm runs, and it can be changed afterwards by the same routes.

The third is the diagnostic. Each discipline leader is asked the firm’s own practices, one at a time, and says whether each happens nowhere, somewhere or everywhere, naming the team wherever it happens somewhere. The result shows where the firm stands, set beside where the sponsor said beforehand it stood, and identifies the teams where the operating model already exists in part. The fourth is use-case selection. We take the use cases the firm already has, pair each with a team the diagnostic identified, score each pair on the team’s delivery record and the use case’s value, and order the pairs in a spreadsheet the firm keeps and can re-rank without us. The top three go to the first workshop day, where the people who own them confirm or amend them. Services and workshops →

The discipline skills

There are nine disciplines, from structure and finance to security and ethics, and 106 practices across them. Each practice is a statement that is either true of a team or not, with a check against the team’s own records. The skills install into the AI tool the team already uses, and they have run in Claude, Codex and Hermes. Each team keeps an overlay of its own, recording the practices it has added and those the sponsor has overruled, so what gets checked is the firm’s practice.

At every commit the skills run the checks and record the result. A practice that fails becomes an open action with an owner, and a policy written as a check can stop a release. Upgrades arrive as versions the team reviews before it adopts them. The nine disciplines, practice by practice →

The digital twin

What the skills find goes into registers kept in your own repositories: who is in each team, who owns what, who decided what, what each team and each agent may do, what the work cost and returned, and which practices hold. Together the registers are a digital twin of the operating model, a record of how the firm works that every release brings up to date. Nothing in it leaves your firm.

Each executive reads the part that is theirs. The chief risk officer sees each exposure with its owner and its age; the finance director sees cost and value to date, with every unpriced line named. A fact nobody has supplied is counted as missing, never shown as zero. The twin across three releases, in a worked example →

The agents

Some steps need judgement rather than a check: answering a diagnostic, proposing a practice, asking an executive’s question, refusing a release. The agent framework takes those steps. Each agent plays its part from a persona, a written description of how the role is held, and the firm can replace any persona with its own. A program controls them: it writes each agent’s instructions, checks what it returns, limits what it may write, and records every act in a log that is only ever added to. The program calls no model itself; the agents run in the AI tool your engineers already use.

At your firm the nine discipline agents act as guidance agents. Each runs its discipline’s skill for the teams it is assigned, and wakes only when their work record changes or an executive asks a question about its discipline, so a day with no change calls no model. Each has an owner, a scope somebody agreed, and a condition on which it stops, and a stopped agent records who stopped it and why. In the simulator the agents are of four kinds, standing in for the firm’s people: one for each role in each team, one for each discipline, one for each executive who reads the twin, and one for the sponsor.

The simulator

The simulator runs the twin forward, with the agents standing in for your people and each team learning at its own rate, and shows how far each discipline moves at every review and what each period costs and returns. Given a policy such as “product owners decide releases under £50,000”, it sets the run with the policy beside the run without it. In the worked example the policy makes work 2.3 days faster from backlog to production and adds three releases, and the share of obligations discharged falls by 0.07. Those figures are the fictitious firm’s. At your firm the simulator runs on a copy of your own record.

Who installs it

Our engineers install it. They set the first targets with the sponsor, install the skills in one team, tune that team’s overlay, and stay until the team runs the skills without them. We hold them to one number: how many teams they bring to that point, and how fast. The firm then adds teams one at a time, and each new team writes into the same twin.

We built the software first for banks, building societies, insurers and asset managers.