KrelunaDocuments and matters
Classify, summarise, search and prepare authorised information.
Kreluna ecosystem
AI and automation for professional services firms
Kreluna Office is being designed to prepare work, highlight missing information and coordinate repetitive activities without replacing professional judgement.
Kreluna is in active development. Early access, feature availability and integration coverage are confirmed individually for each request.
Operational support for professional teams
Email and deadlines
Identify priorities and organise items that require review.
Approval centre
Important actions require confirmation, keeping professionals in control.
From client intake to final approval
Suitable workflows may include organising incoming documents, identifying missing information, preparing draft communication, tracking matters and deadlines, and routing work to an authorised reviewer.
- Document-heavy work in one flow
- Visible ownership and handoffs
- Review before consequential action
Professional judgement stays with the professional
Kreluna prepares and coordinates work; it does not replace specialist or legally regulated advice or professional conclusions. Connection options vary by provider and country and are confirmed before activation.
- Provider-authorised connections where available
- Role-based access and approvals
- A defined early-access use case
Frequently asked questions
Does Kreluna replace professional judgement?
No. It can prepare and organise work; advice, review and professional conclusions remain with the firm.
Which workflow should a firm start with?
Choose frequent, reviewable work such as document intake, missing-information checks or preparation for approval.
Can it connect to an existing practice system?
That depends on the provider’s supported connection methods and is confirmed before activation.
Built around control and clarity
Every project starts with a defined objective, authorised information and a clear review process. Capabilities are enabled progressively and important actions remain subject to approval.
- Defined scope and responsibilities
- Human review for important actions
- Clear limits and measurable outcomes
Explore professional-services workflows
AI for accounting firms
Client documents, missing-information checks and professional review.
AI for law firms
Matters, sources, drafts and lawyer-controlled decisions.
AI for payroll providers and HR consultancies
Employee documents, deadlines and restricted access.
Kreluna resources
Guide: protect confidential data when using AI
Information classification, provider review, permissions and human judgement.
Practical applications
Shared workflows, profession-specific accountability
Accountants, lawyers and payroll advisers share document-heavy work, but their duties differ. Kreluna starts with the firm’s workflow rather than assuming one generic model fits all professions.
Client and matter intake
Collect initial information, classify attachments and prepare a missing-items checklist without deciding whether an engagement should be accepted.
Source-linked files
Find facts and steps in authorised documents and show the passage for checking instead of returning an unsupported answer.
Email and deadlines
Identify requests and dates, suggest tasks and owners, and keep confirmation and sending under the firm’s control.
Draft and review
Prepare summaries or drafts from approved templates with clear status, source, version and reviewer.
Project method
A pilot bounded by matter, team and data
The first test should avoid entire archives and highly unusual cases. Select one workflow, a small user group and an outcome that a professional can check.
- Define the outcome
Describe the problem, who performs the work today and which observable result should improve.
- Map data and exceptions
Identify sources, permissions, manual hand-offs, unusual cases and points where a person must decide.
- Test a narrow scope
Run the pilot on authorised sample data against acceptance criteria agreed before the test.
- Measure and decide
Compare the workflow with its baseline, then extend, revise or stop it on evidence rather than enthusiasm.
Data and accountability
Confidentiality, disclosure and professional control
AI remains an instrument that supports work. The firm defines permitted information, review and client communication, involving legal or privacy specialists where needed.
- Logical separation between clients, matters and roles
- No confidential documents in an initial enquiry
- Visible sources and citation checks before use
- Mandatory approval for client communications and professional work
- Internal rules for permitted tools, retention and incidents
Measuring value
Measures that fit professional work
Speed matters only alongside completeness and verifiability. A good pilot makes errors and responsibility easier to see, not just text faster to produce.
Complete files
Matters reaching review with required information and documents available.
Preparation time
Time spent collecting, classifying, summarising and setting up the work.
Reviewer corrections
Changes needed before a draft can be relied upon or sent.
Traceability
Ability to reconstruct source, version, owner and approval for material steps.
Before you begin
Before sharing a client document
Can we begin with our entire client archive?
That is not the recommended starting point. A minimised, authorised sample makes permissions, quality and system behaviour easier to test.
Must clients be told?
That depends on profession, engagement and use. The firm must assess applicable duties and communicate clearly; Kreluna does not replace competent advice.
Which practice systems can be connected?
Only systems with suitable official interfaces and permission models. An integration is not described as available until it has been verified.
References for responsible design
The AI Act and GDPR can apply together: duties and safeguards depend on the organisation’s role, the data and the system’s actual use. Assessment therefore comes before configuration.
- European Artificial Intelligence ActOfficial text of Regulation (EU) 2024/1689.
- EDPB Opinion 28/2024 on AI modelsAnonymity, legal basis and consequences of unlawful personal-data processing.
- EDPB — Data protection by design and by defaultSME guidance on safeguards, minimisation and accountability.
- OWASP — LLM Prompt Injection PreventionRisks and defences for AI applications connected to documents and tools.
These sources help frame the work; they do not replace legal, privacy or security advice for a specific situation.