Logo

Educational Platform with AI Mentor and Development System

Industry:
EdTech
Timeline:
MVP — 6 months, in progress
Type:
Online Platform
Platform:
Web

A tool for mastering digital skills with personalized learning paths, smart support, and a social component

[project idea]

/01

The goal is to create an alternative to complex, overloaded educational platforms, making learning flexible, intuitive and motivating. No imposed deadlines, with a focus on practice and real skill upgrades.

The foundation is a combination of AI, visual progression, and social support.

[about the project]

/02
Upgrade Paths and Development Goals

Upgrade Paths and Development Goals

The user selects already mastered skills, and the system suggests growth vectors based on them. Each vector corresponds to a set of courses, tasks, and steps.

AI Mentor as a Tutor

AI Mentor as a Tutor

The neural network reviews homework, answers questions, and helps navigate difficult steps. The mentor isn't just a chatbot but a full-fledged learning participant — it maintains dialogue, adapts to context, and can explain complex topics.

Contextual Interaction Between Students

Contextual Interaction Between Students

Within the platform, a user connection system operates: if several people are studying the same topic, they may be offered communication in a micro-group.

[our responsibility]

/03
Product concept and architecture
Development of the upgrade map logic
AI mentor: dialogue design, integration and context-aware responses
UX/UI design for the web platform and Telegram version
Backend, frontend and AI integrations
Content model for courses and modules
Post-launch support

[where we started]

/04
Platform idea focused on career upgrades
Vision of key features: AI mentor, map, micro-groups
Draft prototypes and several test courses
List of tags and skills for MVP

[team]

/05
1 Product Architect
2 UX/UI Designer
3 Backend Engineer
4 Frontend Engineer
5 AI Engineer
6 QA Specialist
[tech stack]
/06
React
React
TypeScript
TypeScript
Next.js
Next.js
WebSocket
WebSocket
OpenAI API
OpenAI API
Python
Python
FastAPI
FastAPI
PostgreSQL
PostgreSQL
Redis
Redis
Docker
Docker

[key features]

/07

Upgrade Map and Path Navigation

AI helps build a map of mastered and target skills. The user sees what they've learned and where they can go next.

Achievement and Motivation System

Badges, progress ribbons, rewards for actions. Everything is designed as a motivational system, without pressure.

Next Upgrade and Global Goal

The system forms the nearest short-term task and captures the long-term career goal. All in clear visualization.

Text and Video Lessons

MVP uses both formats: short text steps and full video modules with interactivity.

AI Mentor Within the Course

Supports, answers, reviews assignments, offers explanations. Integrated into dialogues, steps, and homework.

Telegram Platform Version

Simplified access to courses, homework, and progression via messenger. The full web version remains the main platform.

[development process]

/08
1 Research: EdTech market analysis, user research, interviews with potential students
2 Prototyping: interface, module structure, upgrade and goal logic
3 AI: designing the mentor, defining reaction types and scenarios
4 AI: designing the mentor, defining reaction types and scenarios
5 Development: modular architecture, adaptive logic, Telegram integration
6 Testing: UX tests, content runs, scenario stabilization
7 MVP release: limited number of courses and features for hypothesis testing

[technical challenges and solutions]

/09
01
Challenge

Implement three-level permissions (role, department, entity type)

Solution

Custom ACL module with caching and flexible configuration

02
Challenge

Improve relevance of AI suggestions and reactions

Solution

Building an intermediate layer with enriched context — routes, roles, statuses, task parameters

03
Challenge

Prevent stuck tasks and implicit blockers

Solution

AI generates text comments explaining the reasons and suggests actions (e.g., 'To continue, you need to...')

[results]

/10
  • Average engagement a step above the EdTech market
    + 15-20 %
  • use the AI mentor in the first 3 sessions
    82 %
  • reach their 'next upgrade' within the first week
    67 %
  • the platform passed internal testing and is ready for content scaling

    Confirmed by post-implementation research