COSMOGENESIS

Where it's used

Four questions, one analysis.

The institutions we speak with usually bring one of these four questions. The same analysis sits behind all four, lineage and business rules derived from the code, so once a system has been read, the next question is answered from work already done.

Four questions, one analysisOne building drawn in thin lines with four windows along its front. Above each window is one of the four questions: regulatory evidence, change and migration, AI readiness, understanding a system. Through every window the same interior shows, and one gold line runs unbroken behind all four: the single analysis of the code that each question is answered from.Regulatory evidenceShow the workings behind areported figure.Change and migrationKnow what a change reachesbefore anyone approves it.AI readinessKnow what your AI is standingon.Understanding a systemThe system only a few peoplestill understand.One analysis of the code, behind every window. Four questions, one analysisOne building drawn in thin lines with four windows along its front. Above each window is one of the four questions: regulatory evidence, change and migration, AI readiness, understanding a system. Through every window the same interior shows, and one gold line runs unbroken behind all four: the single analysis of the code that each question is answered from.Regulatory evidenceShow the workings behind areported figure.Change and migrationKnow what a change reachesbefore anyone approves it.AI readinessKnow what your AI is standingon.Understanding a systemThe system only a few peoplestill understand.One analysis, behind every window.

Which of these is yours?

Regulatory evidence. How is this figure actually produced?

When a regulator or an auditor asks how a number is produced, the answer is usually put together by interviewing the people who think they know, and it starts going out of date the day it is signed. Cosmogenesis follows the figure from its source to the report, field by field, with every business rule on the way written in plain English and the code it came from attached. That is the transformation documentation your data standards ask for, produced from the code itself. It identifies which of your data is critical under your own definitions, and produces the same evidence for each of those elements. When the code changes, the evidence is produced again from the new code instead of being re-attested by hand.

Change and migration. What will this change reach?

Before a release, your change board sees which reports, controls and figures it can touch. Before a replacement programme commits, the analysis establishes what the old system actually decides, rule by rule, so the new one is built to match. After cutover, it reads both and compares what each decides, without touching the data. Anything the old system decided that the new one does not is listed, so the old system is retired on evidence.

AI readiness. What is our AI standing on?

Models and agents are being pointed at data whose meaning nobody has fully written down. Cosmogenesis supplies that meaning from the code. Each field arrives with the rules that produce it, the sources it comes from and a statement of what was read, and any model you approve can read it through the API. None of it is written by a model, so it is the same every time you ask. You decide which fields a model may use, and you decide it on evidence.

Understanding a system. Who still understands this system?

Every institution has one. It is critical, decades old and documented mostly by memory. Cosmogenesis derives what it does from its code, including every business rule in plain English, and produces documentation your teams keep using after the people who remember it have moved on. Where the original authors left notes in the code, they are shown beside the analysis, marked as theirs and unchecked. What the code alone cannot explain is listed, which is the part worth asking your longest-serving people about while they are still here.

  • Keeps a row only when ValuedExposure BookClass is 'Banking'. WHERE s.[BookClass] = 'Banking';
  • Calculated as NetExposureBase times RiskWeight. (DT_NUMERIC,19,4)([NetExposureBase] * [RiskWeight])
  • Calculated as the total of BankingCreditRisk CreditRwa. CAST(SUM(s.[CreditRwa]) AS decimal(19,4)) AS [CreditRwa],
  • Calculated as CreditRwa times 0.08. (DT_NUMERIC,19,4)([CreditRwa] * 0.08)
  • Calculated as the number of PositionException rows. CAST(COUNT_BIG(*) AS bigint) AS [ExceptionCount]
  • Calculated as (MarketValue ignoring its sign) times FxRate. (DT_NUMERIC,19,4)(ABS([MarketValue]) * [FxRate])
  • Keeps a row only when ValuedExposure BookClass is 'Trading'. WHERE s.[BookClass] = 'Trading';

Synthetic sample, names as written in its code. No customer data.

In practice

One platform, mapped in full.

    1. Every rule, in plain English
    2. Open one rule
    3. The code behind it
    1. The whole estate
    2. Open one area
    3. A critical field, and why
    1. Trace a field
    2. Where it comes from
    3. How it is worked out
    4. See what depends on it
    1. Documentation you can share
    2. Every field in one place
    3. Find one table
    4. Open one field
    5. Critical data identified

Real screens from the software. Three are of the platform described here, with its name replaced, and the map is of a public sample.

The owner of a technology company had run its platform for twenty years. Its logic sat in four places, a C# and .NET layer, ASPX web pages, JavaScript in the browser and T-SQL procedures in the database.

Cosmogenesis read all four, 698 pages, programs and procedures, and mapped the system from its web pages down to its database. It recorded 7,427 fields, wrote 3,414 business rules in plain English and identified 55 critical data elements. The owner saw how the whole system fits together, found behaviour in it that nobody had known was there, and used the map to plan a refactor.

The foundation

One analysis behind every answer.

Works with what you already haveThree layers. At the bottom, the code your systems run. Above it, Cosmogenesis as a foundation band, an account of what the code does kept current as the code changes. Above that, drawn as equals, the four tools it feeds: Collibra as your catalogue and governance, where ownership, policy and approvals stay; the open standards OpenLineage and OpenMetadata; AI models and agents through the API; and your own tools through the API and files.CollibraYour catalogue and governanceOwnership, policy and approvalsstay here.OpenLineage andOpenMetadataOpen standardsAI models and agentsThrough the APIYour own toolsThrough the API and as filesCosmogenesisAn account of what the code actually does, kept current as the code changes.The code your systems run Works with what you already haveThree layers. At the bottom, the code your systems run. Above it, Cosmogenesis as a foundation band, an account of what the code does kept current as the code changes. Above that, drawn as equals, the four tools it feeds: Collibra as your catalogue and governance, where ownership, policy and approvals stay; the open standards OpenLineage and OpenMetadata; AI models and agents through the API; and your own tools through the API and files.Collibra, your catalogue and governance.Ownership, policy and approvals stay here.OpenLineage and OpenMetadata, openstandards.AI models and agents, through the API.Your own tools, through the API and as files.CosmogenesisWhat the code actually does, kept current as itchanges.The code your systems run

Your catalogue and governance tools hold what has been harvested from your systems and what people have entered about your data. Cosmogenesis supplies what the code actually does and decides, and keeps it current as the code changes. Each question you answer adds to the same analysis.

See it on one of your systems.