Live in production
Say hello
Say hello
to DINA.
Your HR Operations AI Colleague, employee #3, on the job today.
Live today · HR Operations
Meet DINA, the AI Colleague who runs HR.
She runs the HR helpdesk, onboarding and offboarding, with every answer cited to your policy and every consequential action waiting for a human. AI owns execution. Humans own decisions.
Live today
An AI Colleague that handles the highest-load work first.
Three HR modules are live in production now. Each one takes a repetitive, high-volume workflow off your team and runs it with citations, approvals, and a full record.
Live
Onboarding
Collects documents, provisions access through approvals, and gets every new joiner ready from day one to day thirty.
Live
HR Helpdesk
Answers employee questions in seconds with cited policy sources, opens a tracked case, and issues signed, stamped letters (salary certificates, employment, experience and NOCs) once a human approves.
Live
Offboarding
Carries the exit cleanly, from clearance and access revocation to the final-settlement handoff, with nothing missed.
3
HR workflow areas live today
2
Languages: Arabic and English
100%
Grounded answers cite a source
0
Actions ship without a human
Get to know DINA.
Always on
Replies on time, never misses a request.
Online now
Every request opens a tracked case, so nothing slips through the cracks.
Day one to last day
She carries the whole employee journey.
- Join
- Grow
- Move on
Every reply has a receipt
Grounded in your policy, cited line by line.
Leave Policy · §4.2Source cited
You stay in command
She proposes. Your people approve.
Hold
Approve
The same work, governed
What changes when an AI Colleague runs the operational layer.
No invented percentages. Just what changes when DINA takes on the execution your best HR people are buried in.
Before · the HR team alone
With DINA
What changes
Requests scattered across email, chat and corridor conversations.
One audited case, opened the moment someone asks.
Nothing is an orphan. Every request has a number and an owner.
Answers vary by who you ask and how they read the policy.
Grounded answers, cited to the exact policy clause.
Consistency you can point to, not hope for.
Approvals chased across inboxes, signatures and reminders.
One approval queue. The approver sees the exact letter.
Decisions stay with humans; the chasing disappears.
Letters retyped from memory, formatting drifting each time.
Drafted from your templates, checked before any human sees it.
Right the first time, in your house style.
"Who approved that, and when?" Then an afternoon of reconstruction.
Every input, gate and action in the Decision Trace.
Answered in seconds, traceable forever.
Governed by design
An AI Colleague you can actually trust with employee data.
Governance is not bolted on top. It is the architecture. DINA runs inside hard boundaries you can inspect and audit.
A human decision gate on every action.
Every meaningful action stops before execution. An authorized person approves, edits, or rejects it. A small set of critical actions can never be set to auto-approve.
Grounded in your knowledge, cited line by line.
Answers and drafts are grounded in your own approved policies and templates, with a clear source-of-truth hierarchy. Content pending review never reaches a draft.
A trace behind every decision.
A complete, time-stamped ledger records every request, the sources cited, each approval, and who did what, so any decision can be traced from start to finish.
Tenant isolation and PDPL-aware data handling.
Row-level security keeps every customer's data strictly their own, and data handling is built around regional privacy law from the ground up.
Built to fit your business
Works in Arabic and English, grounded in your own policies.
We are launching in the UAE first, so Arabic is supported from day one. DINA understands and replies in Arabic, cites the exact Arabic clause from your policy, and drafts correct right-to-left letters. English and Arabic are both first-class.
Just as important, DINA does not come with one country's rulebook baked in. She grounds every answer in your company's own approved policies, so she reflects your rules and your market, whether you operate in Dubai, Riyadh, or wherever you grow next. Based in Dubai, built to scale globally.
The full HR surface
Phase one is live.
Phase one is live.
Then all of HR.
DINA starts with the three highest-load modules, then expands across the full fifteen-module HR surface, before any other AI Colleague is built.
3 of 15 modules live in Phase 1
12 in design & on the roadmap
M01HR Planning
M02Talent Acquisition
M03Onboarding
M04Employee Experience
M05Payroll
M06Benefits & Wellbeing
M07Records & Service
M08Time & Attendance
M09Performance
M10Learning & Training
M11Compensation
M12Employee Relations
M13Analytics & Reporting
M14Offboarding
M15Immigration & Compliance
Phase 1 · live
Design-only v1
Later
Questions, answered
Frequently asked questions.
What is an AI Colleague for HR?
An AI Colleague for HR is an AI Agent that performs HR operational work, not just chats about it. Unlike an HR chatbot, it triages requests, drafts letters and documents, runs onboarding and offboarding steps, routes approvals to the right person, and records every action. DINA executes the work while consequential decisions stay with your people.
What does DINA actually do?
DINA runs three live HR modules today: an HR helpdesk that answers employee questions with cited policy sources, onboarding from day one to day thirty, and offboarding from clearance to final settlement handoff. She also issues signed, stamped letters (salary certificates, employment, experience and NOCs), opens an audited case for every request, and routes approvals to a named human before anything ships.
How is DINA different from an HR chatbot?
A chatbot answers questions. An AI Colleague completes work. DINA opens a tracked case, grounds every answer in your approved policy with a citation, drafts the document, pauses at a human approval gate, and only acts once a person approves. Every step is recorded in a decision trace.
Does DINA make decisions on its own?
No. NURA AI is built on a single principle: AI owns execution, humans own decisions. Every consequential action pauses at a human approval gate where an authorized person approves, edits, or rejects it. A fixed set of critical actions can never be set to auto-approve.
Does DINA work in Arabic?
Yes. We are launching in the UAE first, so Arabic is supported from day one. DINA understands and replies in Arabic, cites the exact Arabic clause from your policy, and drafts correct right-to-left letters. She grounds every answer in your company's own approved policies, so she reflects your rules and your market rather than any single country's law.
How does DINA keep answers accurate and compliant?
Every answer is grounded in your own approved policies and cited to the source clause. Knowledge is layered across global, jurisdiction, and customer levels and scoped so DINA never speaks past what she is allowed to know. When she does not know, she says so and surfaces the gap for your team to close rather than guessing.
How long does it take to deploy DINA?
About 45 days from kickoff to a production AI Colleague, configured to your SOPs, approval routings, and knowledge. It is a focused implementation, not a multi-quarter program.
How do employees reach DINA?
Employees reach DINA through the channels they already use: email, WhatsApp, Microsoft Teams, and a web request interface. Every request opens a tracked case, and any consequential action is approved by a person before it is carried out.
Is our employee data secure and private?
Yes. Every customer is isolated with row-level security, data handling is built around regional privacy law including PDPL, and every request, citation, and approval is recorded in a complete, searchable audit trail and approval ledger.
Founding Customer Program · Open now
Meet the AI Colleague who runs HR.
Book a 30-minute demo and watch a real employee request flow from message to audit ledger, or message us on WhatsApp and we will set it up.
Reach us at connect@nuraai.ai, a NURA AI specialist replies personally.
