Week1

Practicum Project #

PALTUS team. Report 1

Project name: PALTUS: Personalized Adaptive Learning & Time Utilization System #

Code repository: #

https://github.com/IU-Capstone-Project-2025/PALTUS

Project Idea #

An AI-powered self-learning planner that helps users create personalized study plans for any topic, using AI-model to generate lessons, structure schedules, and track progress with gamification to maintain engagement. User can add a course using AI-model interaction: user writes that he wants to learn some discipline, adds amount of lessons and available time, then AI-model generates a full course depending on user’s preferences and requested topic. The main goal - courses are built to fit user’s comfort and free time. Lessons include an option to edit the course model or lesson topics, add notes on each lesson and see the description of the generated course. Everything is customisable individually, starting from lesson amount and lesson duration, ending with calendar dates and time. There will be an option to give feedback to in a chat with AI-model after lesson or a course.

Problem Statement #

Many students and self-learners struggle to structure their self-study process, leading to inconsistencies and inefficiencies. Existing platforms lack personalization and flexibility, often resulting in lost motivation.

Team Members #

Team MemberTelegram AliasEmail AddressTrackResponsibilities
Sergey Knyazkin (Lead)@poeticlamas.knyazkin@innopolis.universityFrontend/Design/DevOpsCreating UX/UI, designing frontend structure, assisting deployment
Aidar Sarvartdinov@aidar_sara.sarvardinov@innopolis.universityBackendCreating overall backend structure
Amir Fayzullin@HoriFa7za.fayzullin@innopolis.universityFullstackDeveloping frontend components, assisting backend code
Ramazan Gizamov@ramzeuusr.gizamov@innopolis.universityDevOps/Tech communicationApplication deployment, report/presentation writing
Igor Dubrovsky@chomosucei.dubrovsky@innopolis.universityBackendWriting logic for GPT interaction
Danil Demin@degradatorusda.demin@innopolis.universityFrontendCreating frontend components and views

Brainstorming #

Ideas during brainstorming

  1. AI Study Planner (Chosen idea) — An AI-powered self-learning planner that helps users create personalized study plans for any topic, using AI-model to generate lessons, structure schedules, and track progress with gamification to maintain engagement.
  2. Smart Scheduler Bot — A Telegram bot that understands free-form user input about upcoming tasks or events, adds them to a calendar, and sends reminders in advance. Could be integrated with Google Calendar API.
  3. MeetDev — A web platform to help developers find pet projects and teammates by stack and experience level. Aimed at beginners looking for practice and teams searching for collaborators.
  4. CarsToBuy — A web service that aggregates reviews, listings from popular marketplaces, and technical data for each car model to help users make informed purchasing decisions.

Brief market research / problem validation #

Application analogues: The Habitica and Life RPG applications allow you to create tasks, receive rewards for completing them, and buy various equipment, pets, and skills for in-app currency. The Todoist app motivates through points for completing tasks Coursebox.ai and MiniCourse Generator generates courses based on the provided materials (videos, docs) Our app combines the possibilities of creating a course using AI and tracking progress and maintaining motivation through game elements.

Basic requirements #

  • AI-generated topic-based study plans
  • Ability to input any study topic
  • Text-based lesson delivery
  • Clear and simple progress tracking system
  • Web-based interface

Target users and their primary needs #

  • Self-learners / students — need structured and personalized study plans.
  • Busy professionals — want to upgrade skills with limited time using adaptive tools.
  • People preparing for exams / new skills — require guided, gamified paths to stay motivated.

User stories #

  1. As a client, I want a course with structured learning materials to self-learn new skills.
  2. As a client, I need flexibility in my learning schedule to study at my convenience.
  3. As a client, I require clear progress tracking to stay engaged and understand my improvement.

Initial scope #

Included in MVP:

  • AI-generated topic-based study plans
  • Text-based lesson delivery
  • Progress tracking

Excluded (future iterations):

  • Calendar integrations
  • Social features

Tech-stack #

Frontend #

  • Vue.js, Vuetify
    Justification: Vue.js offers simplicity and reactivity. Vuetify provides ready-to-use, well-designed components, accelerating development and ensuring a clean, consistent UI.

Backend #

  • Java Spring Boot
    Justification: Selected for robustness, scalability, and a strong ecosystem for RESTful APIs. Includes powerful tools for authentication, scheduling, and database interactions.

Weekly commitments #

  1. Sergey Knyazkin (lead):
    • Configured Docker for frontend
    • Added Frontend configuration (boilerplate)
    • Created design in Figma
  2. Aidar Sarvartdinov:
    • Created overall idea and most functions
    • Added backend configuration
  3. Amir Fayzullin:
    • Created a prompt for AI-model
  4. Ramazan Gizamov:
    • Wrote report
    • Helped with design in Figma
  5. Igor Dubrovsky
    • Explored available LLM models and their API to utilize them in the project
    • Configured Docker for backend
  6. Danil Demin
    • Suggested format for JSON used in prompt