Role-based AI agents
- Ready-made agents for admissions, student services, financial aid, and IT support
- Set each agent's personality, tone, and when it should hand a conversation to a person
- Answers are grounded in your programs, deadlines, and policies, and the agent is instructed to say it does not know rather than guess
Multi-channel coverage
- A chat widget for your website, with both text and voice
- Email, with proper reply threads and signatures
- Instagram and Facebook: direct messages, comments, and private replies, with safe limits on what is shared in public; WhatsApp through Meta's official Business Cloud API, on your own number
Knowledge ingestion
- Upload your handbooks, prospectuses, and fee tables in any common format
- It reads scanned pages and Arabic image documents, not just typed text
- It can learn directly from your website and keep itself up to date
Connectors and integrations
- Connects to your student system, Postgres, MySQL/MariaDB, or Oracle, to answer personal questions without changing any records
- Personal details stay hidden, and your team approves exactly what it can see
- Actions like a registration request ask for a fresh one-time code, and you can switch them off instantly
Voice-capable chat
- Speaks and understands both Arabic and English
- Students can type or talk, and the agent can reply in voice or text
- Identity verification works in voice too, spoken one-time codes are understood
Smart routing
- Senses stress and escalates automatically when a conversation turns negative
- Escalations land in a shared queue your team works from, with the full conversation attached
- A live queue lets staff take a conversation over and answer it themselves
Approval workflows
- Send low-risk, factual answers automatically
- Hold sensitive topics, like financial-aid decisions, for a person to review and edit before sending
- Test replies in a safe dry run before anything goes out
Analytics for institutions
- Your reply volume and response time
- Sentiment trends for each channel and each agent
- What the AI chose to do, and how often it hands off
Privacy, audit, compliance
- Every stored credential is encrypted, and administrator passwords are hashed
- A tamper-evident log of every action, sealed daily
- A complete consent record, with data export or erasure served from the dashboard on request
Identity verification, when it is needed
- Public information is answered freely, with no sign-in needed
- Personal records like grades, schedule, balance, and application status need identity confirmation first
- Sensitive actions ask for a fresh one-time code every single time