Skip to content
Fully AI-powered HRMS · six agents, fifty-nine tools, twenty-three watchers

The HRMS that runs HR with you, from ten people to ten thousand.

Thirty-five modules on one employee record: people, attendance, leave, shifts, timesheets, payroll, expenses, hiring, performance, learning, assets, helpdesk, policies, engagement, reports and approvals. An assistant that answers from your data, proposes actions you confirm, and watches for problems overnight — under the same roles and scopes as every screen.

Ten demo accounts, one password, every module seeded. No sign-up.

app.dinohrms · Dashboard

Good morning, Priya

Saturday, 6 September

⌘J Copilot

Headcount

248

+6 this month

Present today

91%

12 on leave

Open approvals

7

3 waiting on you

Payroll run

Sep

Review · 2 findings

Attendance, last 12 weeks

avg 82%

AI insights · this morning

  • Payroll pre-lock: 2 employees have no bank account on file.
  • Design team is 40% on leave next Thursday.
35
modules in one sidebar
59
assistant tools, every write confirmed by a person
23
watchers running without a model call
14
seeded roles with scoped permissions

Small and large companies

One product. Sized by configuration, not by edition.

A ten-person startup and a multi-entity group run the same code. The difference is how many companies, branches, roles and workflows you switch on.

10 – 100 people

Small companies

One admin, no HR department, a system that does the HR desk for you.

  • Set up in an afternoon: one company, one branch, fourteen ready-made roles
  • The assistant answers policy questions, applies leave and explains payslips so the founder does not have to
  • Payroll with statutory splits and PDF payslips, no spreadsheet
  • Watchers catch missing punches, unsigned policies and stalled hires overnight
  • Runs on one server with MySQL; no Redis, no message broker, no extra services to babysit
100 – 1,000 people

Growing companies

Managers approve from one inbox; HR stops being the bottleneck.

  • Multi-level approval workflows with conditions and escalation, configured by admins
  • Recruitment pipeline with resume parsing, duplicate detection and one-click hire conversion
  • Onboarding and offboarding checklists that start themselves and chase their owners
  • Goals, review cycles and calibration, with feedback drafts already waiting for managers
  • Dashboards per role and a morning AI digest so leaders see problems before they escalate
1,000+ people

Large companies

Several legal entities, branches and countries under one login, all of it audited.

  • Multi-company tenancy with per-company payroll formulas, calendars and configuration
  • Scopes per permission: self, team, department, branch or company, enforced in the API
  • Searchable audit trail for every edit, approval, login and AI proposal
  • Authenticator-app MFA, session revocation, hardened headers, scheduled restorable backups
  • Bring your own model, cap usage per role, and read the trace of every assistant run

AI in every module

Not a chatbot bolted on. An assistant that knows who is asking.

Questions are routed to one of six specialists. Reads go through the same scope filters as the pages. Writes become proposals that a person confirms, and are executed as that person. Twenty-three watchers run every half hour without a model call, so they never hit a quota.

  • Bring your own model: OpenRouter, OpenAI, Groq, Together, Mistral, Gemini or local Ollama
  • Compensation redacted unless the caller may see it; prompt-injection guard on retrieved documents
  • Every run traced: route, tools, hand-offs and verification, replayable by an auditor
  • Memory per user and per company, with a page to read and delete what it remembers
Explore the AI platform

Ask

Natural-language questions routed to the right specialist. Retrieval over policies, help articles, jobs and past tickets, reranked and cited.

  • Streaming answers with charts and tables attached
  • Memory per user and per company
  • Cached knowledge answers to stay within provider limits

Do

Fifty-nine tools cover leave, attendance, payslips, tickets, approvals, candidates, goals, courses, reports and more. Writes become proposals you confirm, executed as you, under your permissions.

  • Autopilot per tool: off, suggest, confirm or auto
  • Decisions and money always require a confirmation
  • Every run traced step by step

Watch

Twenty-three deterministic watchers run every half hour and after key events: missing punches, leave coverage, payroll audit, SLA breaches, stale goals, hiring stalls.

  • Findings linked to the record they concern
  • Morning digest per person
  • No model call needed, so they never hit a quota

HR helpdesk agent

Answers policy and process questions, citing the help centre and published policies.

Self-service agent

Leave balances, payslips, attendance and requests for the person asking.

Manager agent

Team coverage, pending approvals and one-on-one briefs for people leaders.

Analyst agent

Charts and tables from safe read-only views, with numbers verified afterwards.

Recruiter agent

Screens resumes, spots duplicates, drafts job posts and moves candidates.

Writer agent

Announcements, review feedback, ticket replies and policy drafts in your tone.

Twenty-three watchers, no model call needed

  • Missing punches
  • Leave coverage risk
  • Payroll pre-lock audit
  • Onboarding kick-off
  • Onboarding check-ins
  • Exit clearance
  • Helpdesk SLA watch
  • Helpdesk auto-draft replies
  • Timesheet reminders
  • Probation reviews due
  • Stale goals
  • Hiring pipeline stalls
  • Policy acknowledgements
  • Attrition risk
  • Overtime and burnout
  • Compensation drift audit
  • Leave about to lapse
  • Roster gaps
  • Requisitions and time-to-fill
  • People data quality
  • Access review
  • Review calibration
  • Document intake

Everything HR

Thirty-five modules. One employee record.

Seven suites cover the whole lifecycle. Each one reads and writes the same profile, so nothing is retyped between hiring, paying and reviewing. Every card below is a real page in the app, with what it does and what the assistant does inside it.

People & Organisation

One profile, from offer letter to exit clearance.

The employee record is the spine of the system. Structure, documents, onboarding and offboarding all hang off it, and every reader sees exactly the tier of detail their role allows.

Full suite page

Employees

/employees

A full profile with eleven tabs and a KYC checklist that shows what is still missing.

  • Overview, employment, compensation, documents, bank, dependents, emergency contacts, history
  • Directory tier for peers, full profile tier for HR — personal data withheld by the API, not the UI
  • Bulk CSV import with row-level validation and a dry run
  • Login accounts live on the employee, with role assignment scoped to what you may grant

AI inside

Ask “who reports to Meera in Pune?” and get the answer through your own scope. A watcher flags profile gaps: missing bank, tax id or emergency contact.

My team

/employees/team

Managers see their subtree and peers, with attendance, leave and approvals in one place.

  • Direct reports and the whole reporting subtree
  • Team calendar of leave and shifts
  • Pending requests waiting on you, one click to decide

AI inside

Manager agent briefs you before a one-on-one: attendance, overtime, stale goals and days since last leave for each report.

Documents

/documents

Employee documents, HR letters and policies, viewed inline without downloading.

  • Letter templates with merge fields, rendered to PDF on demand
  • Offer, appointment, confirmation, experience and relieving letters
  • Issued-letter register scoped to who may write, not just read

AI inside

Document intake watcher reads uploads, classifies them and fills the KYC checklist; the writer drafts letters from merge fields.

Organization

/organization

Companies, branches, departments, teams, designations and grades.

  • Multi-company tenancy with per-company configuration
  • Holiday calendars and workweeks per branch
  • Reporting lines that drive approvals and visibility

AI inside

Roster gaps and access-review watchers use the org tree to find teams with no shift and roles nobody should still hold.

Onboarding

/onboarding

Templated checklists that kick off the moment a hire is converted.

  • Task templates by role and location
  • Owners across HR, IT and the manager
  • Day-30 and day-90 check-ins scheduled by an AI watcher

AI inside

Onboarding kick-off opens the checklist the moment a hire is converted; check-ins at day 30 and 90 are scheduled automatically.

Offboarding

/offboarding

Resignation, notice period, clearance and final settlement in one flow.

  • Employee-raised resignation with manager and HR approval
  • Asset return, knowledge transfer and access revocation clearances
  • Full and final settlement handed to payroll

AI inside

Exit clearance watcher chases open assets, access and knowledge transfer; the writer drafts handover notes and exit-interview questions.

Attendance, Leave & Time

Time tracked once, computed nightly, argued never.

Punches, shifts, leave and timesheets feed one ledger. Late marks, overtime and balances are calculated by rules you configure, so payroll starts from numbers everyone already agrees on.

Full suite page

Attendance

/attendance

Web check-in, device imports and manual marks, reconciled against the shift.

  • Daily attendance board with late, early, absent and half-day rules
  • Regularisation requests with approval and audit trail
  • Overtime computed per shift policy

AI inside

Missing-punch watcher spots a punch-in with no punch-out and proposes a regularisation you confirm. Overtime and burnout watch flags people running hot.

Leave

/leave

Leave types, accruals, carry-forward and a ledger that always balances.

  • Policy per leave type: accrual, carry-forward, encashment, sandwich rules
  • Requests with multi-level approval and coverage checks
  • Team calendar and balance statements

AI inside

Apply leave by chat. Leave-coverage watcher warns managers when a team drops below cover, and leave-about-to-lapse reminds people before year end.

Shifts

/shifts

Shift definitions and rosters, including night and rotating patterns.

  • Grace periods, breaks and overtime thresholds per shift
  • Weekly rosters by team or branch
  • Swaps and overrides with approval

AI inside

Roster-gaps watcher finds employees with no shift for the coming week and tells the manager who owns the roster.

Timesheets

/timesheets

Weekly timesheets by project and task, submitted and approved in one step.

  • Project and task lines with hours per day
  • Submission, manager approval and reopen
  • Reminders sent by a watcher before the week closes

AI inside

Timesheet-reminder watcher nudges anyone who has not submitted before the week closes; managers see the gaps in one message.

Payroll & Money

Country-aware payroll with formulas you can read.

Salary structures, statutory splits, loans, expenses and payslips run on a formula builder you configure per company. Every run is prorated, audited and locked before it pays.

Full suite page

Payroll

/payroll

Components, structures, runs and adjustments with a full audit per run.

  • Salary components with a formula builder and statutory splits
  • Runs prorated for joiners, leavers and unpaid leave
  • Tax deducted at source capped at what the period actually paid
  • Calculate, review, lock and pay — with an AI audit tab on every run

AI inside

Pre-lock audit scans every run for outliers before it locks: sudden jumps, zero-pay rows, missing bank details. Compensation drift audit spots pay outside its grade band.

Payslips

/payslips

Employees view and download their own payslips the day the run is locked.

  • PDF payslips viewed inline
  • Year-to-date summaries
  • Ask the assistant to explain any line

AI inside

“Why is my take-home lower this month?” The self-service agent explains each line from your own structure, with figures redacted for anyone else.

Loans

/payroll/loans

Employee loans and advances repaid through payroll automatically.

  • Approval workflow before disbursement
  • Repayment schedule netted from each run
  • Early settlement and holds

AI inside

The assistant answers repayment questions from the schedule and shows the balance after the next run.

Expenses

/expenses

Claims with receipts, approvals and payout through payroll or finance.

  • Categories with limits and policies
  • Manager and finance approval
  • Reimbursement tracked to payout

AI inside

Ask for your pending claims or a team spend summary; the analyst agent returns a table from safe read-only views.

Recruitment, Performance & Learning

Hire, grow and review with the same record.

Requisitions flow into a candidate pipeline, offers become employees, and goals, reviews and courses follow them through their career. Nothing is retyped between stages.

Full suite page

Recruitment

/recruitment

Requisitions, job posts, a kanban pipeline, interviews and offers.

  • Requisition approval before a job goes live
  • Stages you configure, drag-and-drop pipeline
  • Resume parsing for PDF and Word uploads
  • Duplicate-candidate detection and one-click hire conversion

AI inside

Recruiter agent parses resumes, screens against the job, finds duplicates, drafts job posts and moves stages after you confirm. Hiring-stall and time-to-fill watchers keep the pipeline moving.

Performance

/performance

Goals, review cycles, ratings and calibrated appraisal summaries.

  • OKR-style goals with progress
  • Self and manager reviews
  • Competency and KPI frameworks
  • Probation reviews scheduled automatically

AI inside

Managers get a first draft of review feedback from goals and notes. Review-calibration watcher shows rating spread per manager; stale-goal watcher nudges owners.

Learning

/learning

Courses, enrolments and completion tracking for compliance and growth.

  • Video and document lessons
  • Assignments by role or department
  • Completion reports

AI inside

Career-path answers come from the competency framework; the assistant can assign a course to a person or a role after you confirm.

Assets, Helpdesk & Engagement

The everyday requests, answered before they pile up.

Laptops, access, tickets, policies and announcements live next to the people they belong to. Employees raise what they need; the right owner sees it the same minute.

Full suite page

Assets

/assets

Asset register with assignment, return and maintenance history.

  • Categories, serials and warranties
  • Assign and recover during onboarding and exit
  • Audit trail per asset

AI inside

Ask what is assigned to whom. Exit clearance and access review watchers make sure nothing leaves with the person.

Requests

/requests

Helpdesk tickets for HR, IT and facilities with SLAs and assignment.

  • Categories routed to the right desk
  • SLA timers and breach alerts from a watcher
  • AI-drafted replies from the knowledge base

AI inside

Helpdesk agent drafts replies from policies and past tickets; SLA watcher escalates tickets nearing breach to the desk owner.

Policies

/policies

Published policies with versioning and acknowledgement tracking.

  • Rich-text or attached documents
  • Acknowledgement campaigns by audience
  • Indexed for the assistant to cite

AI inside

Every policy is indexed for retrieval, so the assistant cites the exact clause. Policy-acknowledgement watcher chases people who have not signed.

Engagement

/engagement

Announcements, surveys and celebrations that reach the right audience.

  • Targeted announcements by branch or department
  • Pulse surveys with AI summaries
  • Birthdays and anniversaries on the dashboard

AI inside

Writer agent drafts announcements in your tone; survey open-text answers are summarised into themes you can act on.

Dashboards, Reports & Notifications

Every role opens to the numbers that matter to it.

Dashboards are built per persona, reports export what you are allowed to see, and notifications reach people in-app, by email and on the mobile task rail.

Full suite page

Dashboard

/dashboard

Headcount, attendance, leave, hiring and payroll signals with AI findings on top.

  • Persona-specific widgets
  • AI insights card refreshed every morning
  • Proposals inbox for actions waiting on you

AI inside

Every morning an insights job turns watcher signals into ranked findings on your dashboard, each linked to the record it concerns.

Reports

/reports

Employee, attendance, leave and payroll reports with filters and export.

  • CSV and Excel export
  • Scoped to the viewer’s permissions
  • Runnable by the assistant as a table attachment

AI inside

Ask the analyst for a report and it runs the same scoped query you would, attaching the table to the chat.

Notifications

/notifications

In-app inbox, email and digests, deduplicated so nobody is nagged twice.

  • Per-event preferences
  • Morning digest from the AI watchers
  • SMTP configured in Settings, no external queue

AI inside

Watcher findings arrive as a single morning digest per person, deduplicated so nobody is nagged twice about the same thing.

Help centre

/help

Forty-three role-filtered articles with screenshots, also indexed for the assistant.

  • Step-by-step guides per module
  • Filtered to what your role can do
  • Cited by the assistant when it answers

AI inside

All forty-three articles are indexed, so “how do I regularise attendance?” gets the steps and a link to the article.

Workflows, Security & Integrations

Configurable by admins, auditable by everyone who should be.

Approval chains, roles, scopes, audit logs and integrations are configuration, not code. Fourteen seeded roles cover the usual org chart; anything else is a few clicks away.

Full suite page

Approvals

/approvals

One inbox for every request waiting on you, whatever module raised it.

  • Leave, expenses, loans, requisitions, resignations, regularisations
  • Decide from the inbox, the record or the assistant
  • Every decision written to the audit trail

AI inside

Decide from the chat: the assistant shows the request and its context, proposes approve or reject, and executes under your permission once you confirm.

Workflows

/workflows

Multi-level approval definitions per entity, with conditions and escalation.

  • Steps by role, manager chain or named user
  • Conditions on amount, days or department
  • Escalation when a step stalls

AI inside

The assistant knows which step a request is waiting on and who owns it; escalation is automatic when a step stalls.

Roles & permissions

/roles

Permission matrix with self, team, department, branch and company scopes.

  • Fourteen seeded roles from Company Admin to Auditor
  • Scope per permission, not just per role
  • You may only assign what you hold

AI inside

AI policy lives here too: which agents and tools each role may use, daily caps and autopilot levels per tool.

Security

/security

Audit trail, sessions, MFA and user administration.

  • Authenticator-app MFA
  • Searchable audit log per actor and entity
  • Session revocation and login history

AI inside

Every assistant run writes a trace: route, tools, hand-offs and verification. Access-review watcher lists role assignments that no longer match a person’s job.

Integrations

/integrations

Connections to devices, email and external systems, with webhooks.

  • Biometric and attendance device imports
  • Outbound webhooks on events
  • API keys with scoped access

AI inside

Device imports feed the same attendance ledger the watchers read, so a missing punch from a biometric device is caught the same way.

Settings

/settings

Company, localisation, email, backups and everything else in one place.

  • Country, currency and date formats
  • SMTP and email templates
  • Scheduled, restorable backups

AI inside

Choose your model provider, set caps and see index health from the AI provider page; everything else the assistant needs is already configured.

Rollout

Live in five steps, not five months.

Everything is configuration. The demo you can open today was built with exactly these steps and a seed script.

  1. 01

    Set up the organisation

    Companies, branches, departments, designations, grades, shifts, holiday calendars and leave policies. Country, currency and formats per company.

  2. 02

    Bring people in

    Import employees from CSV with row-level validation, or hire through the recruitment pipeline and convert an offer into a profile in one click.

  3. 03

    Assign roles and scopes

    Fourteen seeded roles cover the usual org chart. Adjust the permission matrix, choose scopes, and decide which AI agents and tools each role may use.

  4. 04

    Run the month

    Attendance and leave feed the ledger, timesheets close, approvals flow through workflows, and payroll calculates, audits, locks and pays.

  5. 05

    Let the assistant work

    People ask it questions, managers confirm its proposals, and twenty-three watchers keep running every half hour with a digest each morning.

Built for every seat

Each role opens to a different product.

Fourteen seeded roles, from Company Admin to Auditor. Navigation, dashboards, reports and the assistant all reshape themselves around what a person may see and do.

Company Admin

Configure companies, roles, workflows, the AI provider and backups. See everything, change anything, all of it logged.

HR Admin

Run people operations: profiles, onboarding, leave policy, letters, engagement and approvals.

Reporting Manager

Approve from one inbox, watch team coverage, write reviews with a draft already waiting.

Employee

Check in, apply for leave, read payslips, raise a ticket and ask the assistant anything about your own record.

Trust

Configurable by admins, auditable by everyone who should be.

Security is not a module you buy later. Roles, scopes, MFA, audit and backups ship in the core.

Scoped visibility

Self, team, department, branch or company — set per permission, enforced in the API, obeyed by the assistant.

Audit on every action

Approvals, edits, logins, sessions and AI proposals are written to a searchable trail per actor and entity.

MFA and hardened headers

Authenticator-app MFA, session revocation, per-request content security policy, no inline scripts.

Country-aware

Countries, states, cities and phone codes seeded; payroll formulas and statutory splits per company.

Multi-company

Several legal entities under one login, each with its own configuration, calendars and roles.

Workflows, not code

Multi-level approvals with conditions and escalation, configured by admins in minutes.

Roles, workflows and security

Deployment

Your server, your data, your model.

One Docker stack, one database, any OpenAI-compatible model. Nothing leaves your infrastructure unless you point it somewhere.

Self-hosted, one stack

Web, API, worker and MySQL in Docker. Optional Qdrant for retrieval. No Redis, no broker: the queue, locks and scheduler live in MySQL.

Backups you can restore

Scheduled backups with a tested restore path, configured from Settings, not from a shell.

Your model, your keys

OpenRouter, OpenAI, Groq, Together, Mistral, Gemini or a local Ollama. Switch providers from the AI provider page.

Email without a third party

SMTP configured in Settings, templates editable, digests and notifications deduplicated.

Country-aware from day one

Countries, states, cities and phone codes seeded; payroll formulas and statutory splits per company.

Every screen responsive

Mobile task rail for approvals and check-ins; the assistant is one shortcut away on any device.

Questions

Straight answers.

The short version of what people ask before a walkthrough. The long version is the demo.

Is the AI optional?

Yes. Every module works without a model. Watchers are deterministic queries and run on any plan. The assistant needs a provider key; pick one from the AI provider page.

Can the assistant see salaries?

Only for callers who hold a compensation permission. Pay figures are stripped from tool output before they reach the model for anyone else.

Can it take actions on its own?

Reads, yes. Writes become proposals a person confirms and are executed as that person. Decisions and money never run on full autopilot, whatever the autopilot setting.

Which model providers work?

Any OpenAI-compatible endpoint: OpenRouter, OpenAI, Groq, Together, Mistral, Gemini, or Ollama running on your own hardware.

Does it handle more than one company?

Yes. Several legal entities live under one login, each with its own configuration, payroll formulas, calendars and roles.

What do we need to run it?

Docker and a server. The stack is web, API, worker and MySQL, with optional Qdrant for retrieval. There is no Redis and no message broker.

How are roles handled?

Fourteen seeded roles from Company Admin to Auditor. Each permission carries a scope, and you can only assign roles you hold yourself.

Is there an audit trail?

Every edit, approval, login, session and AI proposal is written to a searchable log per actor and entity. Assistant runs are traced step by step.

See it run with real data.

Every module is seeded — ten demo accounts, one password, a live walkthrough that survives clicking anything.