Client
Kaizen Club
Selected work / Kaizen Club
Kaizen Club · TheNext
AI infrastructure for a three-day business summit in Abu Dhabi — one Telegram Mini-App from the first ticket click to materials after.
Platform
Telegram Mini-App
Build window
7 weeks
Kaizen Club
Telegram Mini-App, sales bot, QR access and smart seating, Kai Tinder networking, AI business diagnostic with PDF, interactive AI avatars, AI talk recaps, custom admin panel.
The work
AI infrastructure for the summit in Abu Dhabi: sales, entry, schedule, networking, AI business diagnostic, and AI avatars — all inside one Telegram Mini-App.
What mattered
The summit was positioned as the first AI-powered one. That means AI in every touchpoint, not one robot on stage. A premium audience notices the gap between promise and reality first.
What shipped
Mini-app, bot, QR access, Kai Tinder, AI diagnostic with PDF, live avatars, AI talk recaps, and a custom admin panel — all running for three days across 200+ attendees.
Case study
Kaizen · TheNext
AI infrastructure for a three-day business summit hosted by <em>Margulan Seissembai</em> in Abu Dhabi — the flagship event of his Kaizen Club. <em>One Telegram Mini-App: from the first click on a ticket to the materials after.</em> Not a chatbot and not a single stage moment — the layer the whole event lives on: sales, entry, program, networking, diagnostics, AI avatars. Seven weeks to build. The venue date did not move.
Problem
If you promise AI everywhere, you have to build it everywhere.
The summit was positioned as the first AI-powered one: AI in every touchpoint, not one robot on stage. A premium audience notices the gap between the promise and what actually opens on the phone first. One team had to close the sales bot, the mini-app with entry and schedule, voice AI avatars, the business diagnostic, seating, and post-event materials. Seven weeks. Venue date fixed.
Solution
One product from the first click to the materials after.
We folded everything into one Telegram Mini-App: one account, one database, one admin panel. The build follows the attendee journey, each stage with its own feature set. A few decisions went into the final beyond the brief: the schedule moved onto the home screen, a shared AI generation feed replaced private video invites, and Face ID at entry was replaced with a personal QR — faster and more stable under peak load.
Technical work
Orchestrating four AI providers under one shell.
Four independent AI providers run inside one app: HeyGen for avatars, ElevenLabs for voice, separate LLMs for the conversational agents, and another LLM chain for the diagnostic. The user sees one interface — we built a thin product layer over several APIs with different limits and response formats. The AI diagnostic is a strict methodology — the AI Maturity Index: six dimensions of company AI maturity, a 1–5 scale on each, and a PDF report with interpretation.
Related work
Almaty Marathon
An AI support agent for one of the largest running events in Central Asia, answering participant questions and routing edge cases.
- Built
- Support agent, event rules ingestion, policy updates, QR and pack-pickup flows, escalation logic.
- Stack
- LLM agent · Knowledge ingestion · Escalation routing
Magnum HR Agent
An AI HR agent for a Kazakhstani retail chain: candidate screening, internal knowledge, vacancy fit, and recruiter workflows.
- Built
- Conversational agent, admin panel, knowledge ingestion, eligibility logic, age logic, notification flows.
- Stack
- Internal HR systems · Infobip · WhatsApp · Multilingual LLM
LiftEd
An AI platform for schools that turns checked work into a clear map of knowledge gaps, progress, and next steps.
- Built
- Work checking, gap diagnostics, knowledge map, adaptive practice, school analytics, and role-based product flows.
- Stack
- LLM grading · Adaptive engine · Role-based UIs · Analytics
Running an event where AI has to live across the whole attendee path?
We help turn the AI promise into a product infrastructure: one interface, several AI modes under the hood, and an admin panel for the event team.