Week1

Week 1 #

Project description #


Project name: TeamSync

Code repository: link

TeamSync is a smart university team-matching platform that helps students quickly and effectively find teammates for academic projects. It streamlines the team formation process through AI-based matchmaking, smart user and project profiles.

Team #


Team memberTelegram aliasInnopolis EmailResponcibilities
Diana MInnakhmetova (Lead)@diana_minnd.minnakhmetova@innopolis.universityProduct management, design, Report writing
Danis Sharafiev@HarneMerd.sharafiev@innopolis.universityML + MLOps
Daria Alexandrova@ae_quord.alexandrova@innopolis.universityFrontend
Stepan Dementev@dementevssstepans.dementev@innopolis.universityBackend + DevOps
Elizaveta Zagurskih@wkwthigoe.zagurskih@innopolis.universityBackend
Kamilya Shakirova@hugecatwithacheburekk.shakirova@innopolis.universityML

Brainstorming #


Ideas during brainstorming #

RankIdea
1Smart university team-matching platform that helps students quickly and effectively find teammates for academic projects
2Dynamic resource allocation between services by means of time series. For example, allocate a little more RAM to one or another service, give more cores, etc. Formalize as SaaS
3AI-powered habit tracking application designed to help users develop and maintain positive habits through personalized recommendations and consistent tracking.

Problem validation #

1. Smart University Team-Matching Platform

Problem: Students spend 5–10 hours forming teams manually, often leading to mismatches and delays.

Validation:

  • Most use unstructured Telegram chats, spreadsheets. One of the more or less same sevice is already not working ( Pet-Plat)
  • Professors report frequent issues with unbalanced teams.

2. Dynamic Resource Allocation SaaS

Problem: Cloud services often over- or under-use resources due to static allocation.

Validation:

  • Time-series-based autoscaling is complex and costly to configure manually.
  • Existing tools like Cast.ai are powerful but too advanced for small teams.

3. AI-Powered Habit Tracking App

Problem: Current habit trackers lack personalization, causing low long-term engagement.

Validation:

  • Popular apps use generic reminders, not behavior-based adaptation.
  • Users want smarter, context-aware support.

Basic Requirements #


Problem Statement #

Students of Innopolis University currently face the inefficiencies and time-consuming manual efforts when forming teams for academic projects, particularly during Capstone and elective-based collaborations. This results in suboptimal groups and delayed project starts.

Solution #

TeamSync aims to improve the quality and balance of student teams by offering a data-driven, user-friendly experience for discovering, evaluating, and matching with potential teammates.

Target Users #

Primary Customers: Innopolis University or other organizations.

Primary Users: Students of Innopolis University.

Secondary Users: Professors.

User Stories #

  1. As a student, I want to create a personal profile with my skills, interests, and past projects so that I can be discovered by project creators or teammates.
  2. As a student, I want to post a new project with a description and required skills, so that I can find teammates who match my needs.
  3. As a student, I want to receive AI-generated teammate recommendations based on my skills and preferences, so that I can build a balanced and effective team faster.
  4. As a student, I want to apply to projects that match my interests, so that I can participate in projects relevant to my skills and goals.
  5. As a project owner, I want to accept or reject team applications and manage my team members, so that I can control who joins my project and maintain a good team structure.
  6. As a student, I want to see suggestions and filters when browsing projects (similarity), so that I can efficiently choose which projects to join.
  7. As a professor, I want to make a page only for my course, so that I can manage the team formation between my students.

Main Functionalities #


Smart Profiles #

A system that allows students to create detailed profiles with skills, interests, availability, and academic background.

  • Students can input relevant information including tech stack, preferred project types, and availability.
  • Students can make tags for better search and matching.
  • The system supports real-time updates as students gain experience or complete courses.

Project Hub #

A centralized interface to browse, create, and manage project listings.

  • Students can post project ideas with descriptions, required roles, and deadlines.
  • Other users can browse available projects, filter by tags or requirements, and submit join requests.
  • Each project shows team composition, required skills, and current progress status.

AI Matchmaking #

An intelligent recommendation engine for finding optimal team compositions.

  • Recommends teammates or projects.
  • Search engine.

Team Management Tools #

Features to organize and manage project teams effectively.

  • Project owners can accept/reject applications and manage current team members.
  • Team dashboards display role distribution, availability conflicts, and overall skill coverage.
  • Notification system for team activity and status updates.

Analytics for Instructors #

A dashboard for faculty to monitor team formation patterns.

  • Includes ungrouped students, skill demand gaps, and team distributions.
  • Helps professors support team-building proactively.

Search & Tagging System #

Enhanced discoverability through structured filtering and semantic tagging.

  • Projects and profiles are enriched with tags based on content.
  • Students can search based on interests, tools, tech stacks, or availability windows.
  • Search results are ranked by relevance and user preferences.

User Feedback & Improvement Loop #

  • Users can rate or flag recommendations to improve accuracy.
  • Users can share feedback to the telegram of this project (will be created soon)

System Architecture #


Link to the Figma Board with system architecture:

https://www.figma.com/board/se6mmoDbUnNADLcPqSszQE/Untitled?node-id=0-1&t=GFqAGJxXii04Q0gK-1

Initial Scope (MVP) #


IN Scope (W1–W3)OUT (Post-MVP)
User auth (signup/login)Team invitation/management
Profile creation (skills, availability)Profile dashboard
Project creationAdvanced analytics dashboard
Search engine
Basic recommendation systemAdvanced recommendation system

Tech Stack #


ChoiceJustification
FrontendReactFor building dynamic and high-performance user interfaces and server-side rendering capabilities.
TypeScriptA strongly typed programming language that builds on JavaScript, providing better tooling at any scale.
Tailwind CSSA utility-first CSS framework for rapid UI development with consistent and responsive design.
BackendJava/Spring BootRobust REST APIs, team expertise
DatabasePostgreSQLACID compliance for user/project data
RedisFast session/matching-cache access
MLPythonFor its simplicity and extensive libraries for data processing and machine learning.
AirflowPipeline orchestration
InfraDocker + Docker ComposeEnvironment consistency, TA reproducibility
GitHub Actions (CI/CD)Automated testing/deployment

Individual commitments #


Team memberTelegram aliasContribution
Diana MInnakhmetova (Lead)@diana_minnWrote report, hold 3 meetings, organized workflow in Jira and Figma, made mind map and user flow, wrote README, made logo and design
Danis Sharafiev@HarneMerMade system architecture, helped with initializing ML
Daria Alexandrova@ae_quorInitialized basic Frontend, made design
Stepan Dementev@dementevssstepanOrganised GitHub workflow, helped with system architecture, made Docker compose, initialized basic backend
Elizaveta Zagurskih@wkwthigoMade ERD diagram, Initialized basic backend
Kamilya Shakirova@hugecatwithacheburekInitialized basic ML

Links #


Confirmation of the code’s operability #

We confirm that the code in the main branch:

  • In working condition.
  • Run via docker-compose (or another alternative described in the README.md).