Review with the source in view
AI proposes structured contract fields with a page and quote. A reviewer verifies material values before they become official records.
Explore contract reviewAI-assisted contract & supplier operations
From contracts
to accountable
records.
Turn unstructured agreements into source-traceable records. Connect human review, rule-driven approvals and the work that follows a signature.
A portfolio project by Selina Armstrong. Fictional data throughout.
The accountable record
01 / Source
A page. A quote.
Know where each proposed value came from.
02 / Review
A human decision.
Keep verification and exceptions explicit.
03 / Register
A traceable record.
Preserve the source, reviewer and history.
Evidence is part of the record.
The review workflow, illustrated.
Source-traceable.
Human-verified.
Rule-driven.
The work beyond extraction
One connected record.
Every step accounted for.
Contract data becomes useful when you can explain it, approve it and act on it. This workbench brings those steps together.
Make approvals explicit
Deterministic rules route approvals and exceptions. Mandatory approvals must be complete before an executed contract enters the register.
Explore approvalsKeep suppliers connected
Connect supplier records to contracts, qualification documents and explainable risk factors, while preserving human-maintained master data.
Explore supplier recordsTurn dates into accountable work
Assign obligations, track renewal dates and record completion evidence. Overdue status follows the due date, rather than a manually chosen label.
Explore obligationsFollow the contract lifecycle
Preserve original terms, amendment changes and effective values. Keep the relationship between documents, versions and operational records visible.
Explore the contract registerEvaluate the AI, too
Inspect field accuracy, source coverage and unsupported values using fictional evaluation fixtures. Seeded demo reports are labeled separately from live model runs.
Explore AI validationThe product principle
AI proposes.
Rules guide.
People confirm.
A confident extraction is still a proposal. An executed record needs supporting evidence, the right approvals and a reviewer who can stand behind the decision.
See the product decisionsBefore you explore
A clear scope.
An honest demonstration.
What is ContractLedger AI?
An AI-assisted contract and supplier operations workbench built by Selina Armstrong as a portfolio project. It demonstrates how source evidence, human review, approval rules and audit history can support reliable operational records.
Who is this project for?
Contract administrators, legal operations teams, procurement specialists and hiring teams evaluating practical work in contract operations and vendor governance.
Does AI make the final decision?
AI proposes extractions and summaries. Deterministic rules control workflow requirements, and people confirm material values and decisions. The project does not offer autonomous legal advice.
Is this a full contract lifecycle management platform?
Its focus is contract administration: review, registration, amendments, suppliers and obligations. Negotiation, e-signature, enterprise identity and multi-tenant operations are outside the current scope.
Can I explore the demo?
The public deployment supports guest browsing when enabled. Sign-in and write capabilities depend on the configured role. All contracts, organisations, people and policies in the demo are fictional.