Palantir "Gotham" Integration & Governance

Palantir "Gotham" Integration & Governance is a Moraph-managed service that builds and runs the data pipelines feeding your Palantir Gotham deployment. Each source is mapped, checked for completeness and consistency, traced from origin to ingestion, and checked for classification and handling markings before it is loaded. Records that fail a check are held for your data stewards and trained classifiers to decide on, so analysts work from data whose origin and markings can be shown.

The problem it solves

Analysis is only as good as the data behind it. When feeds are loaded into an analytic platform without consistent checks, duplicates, missing fields, mismatched identifiers and stale records flow straight into the picture analysts rely on. Classification and handling markings may be missing or inconsistent between sources, which risks both over-restriction and spillage. And when a finding is questioned, it can be hard to show where each piece of data came from and what was done to it on the way in.

Who it’s for

• Intelligence and defense organizations that run Palantir Gotham
• Data owners and stewards responsible for the sources that feed it
• Security and classification management staff responsible for markings
• Analysts who need to trust and explain the data they work with
• Oversight, compliance and inspector general staff who review data handling

1. Onboard a sourceEach new source is authorized by your organization, profiled by Moraph, and mapped to the Gotham data model you have approved, with quality and marking rules agreed before any data is loaded.
2. Check and loadOn every run, records are validated, de-duplicated and checked for required markings before loading. Records that fail are held, not loaded, with the reason recorded.
3. Resolve held recordsData stewards and trained classifiers work the held records: fixing them at the source, correcting markings under your procedures, or rejecting them, with each decision recorded.
4. Trace and reviewFor any record in Gotham, lineage shows its source, the pipeline version that loaded it, the checks it passed and any decisions made on it, for analysts and oversight staff.

Managed pipelines

Moraph builds, runs and maintains the pipelines from each authorized source into Gotham, under version control.

Source profiling

Each new source is profiled for fields, completeness, identifiers and markings before mapping begins.

Data model mapping

Source fields are mapped to the Gotham data model your organization has approved, with the mapping documented.

Quality checks

Required fields, formats, duplicates and references checked on every run, with rules your data stewards approve.

Marking checks

Records are checked for required classification and handling markings and for consistency between record-level and field-level markings.

Hold, don't load

Records that fail a check are held with the reason rather than loaded, so problems are fixed before analysts see them.

Steward and classifier queues

Held records go to the right people: quality failures to data stewards, marking failures to trained classifiers.

End-to-end lineage

Every loaded record carries its source, extraction time, pipeline version, checks passed and decisions made on it.

Access mapping

Markings and source rules are mapped to the access controls you approve in Gotham, and the mapping is reviewed with your security staff.

Pipeline monitoring

Moraph watches runs, volumes and failures, and raises issues before stale or incomplete data reaches analysts.

Change control

Pipeline, rule and mapping changes are raised, tested, approved through your process and recorded.

Suggested fixes, human decisions

Where a rule suggests a likely correction, such as a missing marking inferred from the source, a person reviews and decides; nothing is changed automatically.

Data owner

Authorizes sources and approves how they are used

Data steward

Approves quality rules and decides on records held for quality

Classifier

Decides on records held for markings, under your classification procedures

Analyst

Uses lineage to understand records and reports suspected data issues

Moraph service manager and data engineers

Build and run the pipelines and checks under your approvals

Oversight reviewer (read-only)

Reviews lineage, decisions and changes without changing anything

Palantir Gotham

Checked records are loaded through the integration interfaces Palantir provides for your deployment, mapped to your approved data model.

Source systems

Authorized sources are connected by file transfer, database extract, message feed or API, as each source allows.

Cross-domain transfer

Where a source sits on a different network, data arrives through the cross-domain solution your organization has approved, including Moraph's Cross-Domain Solution (CDS) Transfer Agent.

Identity provider

Moraph's governance tools are reached through your identity provider, with groups mapped to roles.

SIEM

Pipeline runs, held records, decisions and administrative actions are sent to your SIEM.

IT service management

Changes, incidents and source requests are tracked in your service management tool, or in Moraph's if you have none.

Data catalog

Source descriptions, mappings and lineage can be published to the data catalog your organization uses.

Single sign-on

Staff sign in through your identity provider, so your authentication policies apply.

Role-based access

Data engineers, stewards and classifiers see only the sources and held records their role and clearance allow.

Encryption and keys

Data in pipelines and holding areas is encrypted, with keys Moraph staff cannot use without explicit, time-bound authorization.

Complete audit trail

Every run, check result, decision and change is logged with the person and time.

Cleared US support staff

Tickets for federal and defense clients are handled only by US citizens with active clearances.

Your authority on classified networks

Pipelines on classified networks are built and run under your organization's authority and procedures; Moraph holds no accreditation of its own.

Alongside your Gotham deployment

Pipelines and governance tools run in the same environment as your Gotham deployment, on infrastructure your organization authorizes.

US-hosted service tooling

Where unclassified service tooling is used, it is hosted in Moraph's United States region only.

Classified networks

Where pipelines run on classified networks, Moraph's cleared staff work under your organization's authority and procedures, scoped in the proposal.

Discovery

We inventory the sources feeding Gotham today, how they are loaded, and the data model and access controls in use. We need your data owners and the team that administers Gotham.

Governance rules

Quality and marking rules are agreed with your data stewards and classification management staff, and documented.

Pipeline build

Pipelines are rebuilt or built new under version control, with checks, holding areas and lineage recording, starting with the sources you choose.

Parallel run

Where existing loads are being replaced, old and new pipelines run side by side and differences are reviewed before cutover.

Queues and roles

Steward and classifier queues, roles and escalation paths are configured to match your procedures.

Transition to operations

Runbooks, monitoring, reporting and service reviews are agreed, and the service moves into steady-state management.

• Moraph is not Palantir. Palantir owns and supports Gotham, and its own service commitments, roadmap and outages apply.
• Licenses for Gotham are subject to Palantir's terms.
• Moraph does not classify information. Marking checks find missing or inconsistent markings; classification decisions are made by your trained classifiers.
• Checks catch the problems their rules describe. They cannot make a source accurate, and data that is wrong but well formed can still pass.
• Moraph does not decide who may see data. Access controls in Gotham are configured as your organization approves.
• The service does not change Gotham itself or its analytic features; it governs the data that goes into it.
• Authority to collect, retain and use each source, including any information about US persons, is your organization's responsibility and is confirmed before a source is onboarded.
• Moraph does not hold accreditation for classified environments; operations there take place under your organization's authority.

• Pick two sources that cause the most analyst complaints and have them profiled
• Review the proposed quality and marking rules with your data stewards and classification management staff
• Trace a sample record from source to Gotham and check the lineage is complete
• Walk a held record through the steward and classifier queues
• Confirm how markings map to access controls in your Gotham deployment
• Read the responsibilities table against your Palantir support arrangements