Alara Ai
Most recently, I’ve been working on Alara within Fort Wise Ai Solutions — building an internal admin panel to manage AI-based operations. This is essentially an admin panel designed to help clients interact with the Alara AI assistant, manage orders, and communicate with their own customers across major platforms — WhatsApp, Telegram, Instagram, Facebook and Viber. Whenever the AI fails to respond effectively, human agents can take over the conversation directly from the panel.
To support this, I built a complex multi-role ecosystem including clients, their team members who operate Alara, system administrators who assist them, and other critical user types. Each role has tailored permissions and responsibilities to ensure smooth collaboration and control.
I also developed the backend infrastructure that powers the panel in real-time, using a configuration I call “4+1” — Alara AI + Messaging Bot + Application Server + Database + Admin Panel.
I am responsible for the entire lifecycle: concept creation, design, development, and deploying the MVP to production. The result is a scalable system that combines automation with human oversight, helping teams stay efficient while delivering seamless user communication across platforms.
- Client
Serenity Transfer Service Ltd.
- Year
2025
-
Scope of work
System architecture for AI admin panel, Multi-role permission system, Back-end development (Node.js, TypeScript), Front-end with real-time updates, Integration with WhatsApp, Telegram & Instagram APIs, MVP delivery, UX design for AI-to-human fallback workflows, Complex messaging bot orchestration.
Development Process Overview
My first task in the project was to create a commercial website to promote the agent. I integrated Alara into the site to demonstrate its capabilities and connected marketing tools for advertising.
Next, I moved on to building an internal admin panel for clients — a tool that allows them to manage Alara, handle orders, and communicate with their customers across WhatsApp, Telegram, and Instagram. I was responsible for designing the entire user experience: I created the UX/UI design, wireframed the structure, built prototypes, and conducted A/B testing.
For the frontend, I used Svelte; for the backend, I built an Express server on Node.js, and developed all necessary endpoints using TypeScript. I implemented the database using Supabase, optimizing its architecture for scalability and fast access.
I also built a multi-role system that supports different user types — clients, their team members, system administrators, and others — each with clearly defined permissions and responsibilities. I integrated webhooks to enable real-time communication between the user, the Alara AI, messaging bots, the admin panel, and the database.
