Week #6

Week #6 #

Links #

Project overview #

Smartify is an educational platform for Russian high school students that helps them choose a profession, choose the right university, prepare for the Unified State Exam, and find a tutor. The main feature is the use of artificial intelligence for personalized recommendations that make the path to admission clear and accessible.

The platform offers four main page. In the career guidance page, students complete a questionnaire indicating their interests, academic performance, and hobbies. Based on the answers, the system suggests specific professions, immediately indicates the necessary subjects for admission and explains why this profession is suitable.

The university selection page allows you to filter universities by region, form of study (budget/fee) and direction. It is possible to compare universities by passing scores, tuition fees and rating.

The Π•Π“Π­ preparation page offers a list of necessary subjects depending on the chosen profession and university, forms an individual curriculum, contains built-in mini-tests and a progress tracker.

The tutor page allows you to search for teachers by subject, teaching style (friendly, strict, etc.), as well as view detailed profiles with information about education and experience.

Since our project is quite large and difficult to fully implement in 2 months, we plan to continue working on it and maintain

Features #

🟨 Frontend #

1. πŸ” Authentication #

Registration (Multi-step) #

  • Email validation with confirmation code
  • 5-digit code input (via pin_code_fields package)
  • Password strength check:
    • Minimum 8 characters
    • Must include digits and special symbols
    • Strength visualization with color indication

Password Recovery #

  • Full cycle:
    • Reset request β†’ code verification β†’ new password setup
  • Data validation at each stage

2. 🧭 Main Interface #

  • Transitions between main sections:
    • Universities, Exams, Careers, Tutors
  • Side menu with SlideTransition animation

Responsive Design #

  • Fixes for overflow issues and text wrapping
  • Clickable cards with visual effects (InkWell, hover animations)
  • Mobile layout optimization

3. πŸ“Š Data Management #

JSON Handling #

  • Parsing:
    • professions.json
    • spheres_stats.json
  • Filtering logic:
    • Spheres β†’ Sub-spheres β†’ Professions
  • Dynamic filtering of professions
  • Search result highlighting

Databases #

  • Initial implementation of the university list

4. πŸ“ Questionnaire #

Dynamic Loading #

  • Parsing of DOCX/XML files using archive and xml packages

Question Types #

  • Single / multiple choice
  • Scales (1–5 / 1–7)
  • Text fields (e.g., β€œOther”)

Validation #

  • Required field checks
  • Smart suggestions
  • Default values for sliders

5. 🎯 Recommendation Page #

Display #

  • Large cards for top 3 recommendations
  • Compact adaptive cards for the rest
  • Sections:
    • Pros
    • Cons
    • Description
  • Color-coded blocks
  • Field mapping using subsphere

6. πŸ—‚οΈ Profession Navigation #

Hierarchical Structure #

  • Spheres β†’ Sub-spheres β†’ Professions
  • Parameterized navigation

Detailed View #

  • Unified ProfessionCard component
  • Section-based content layout
  • Bulleted lists for key characteristics

7. βš™οΈ Technical Solutions #

State & Animations #

  • State management via StatefulWidget / setState()
  • Smooth transitions, hover animations

Optimization #

  • Data preloading
  • JSON caching
  • Dynamic lists via ListView.builder

8. 🌐 Localization #

  • app_localizations_en.dart: English translations added
  • app_localizations_ru.dart: Russian translations added

🟦 Backend #

1. πŸ” User Authentication and Login #

  • Secure sign-in and authentication flow for users

2. 🧭 Data Storage in MongoDB #

  • Persistent storage of specific user or app data using MongoDB

3. πŸ“Š Data Storage in PostgreSQL #

  • Relational data storage for structured information

4. πŸ“ Access and Refresh Tokens #

  • Token-based authentication for secure session management

5. πŸ—‚οΈ Local Data Storage with SharedPreferences #

  • Lightweight key-value storage for client-side data persistence

6. 🎯 Secure Token Storage on Client #

  • Storing access and refresh tokens using secure storage mechanisms

7. βš™οΈ Client-Side University Data as Files #

  • University data stored locally on the client as static files

8. 🌐 Multiple API Endpoints for Communication #

  • RESTful endpoints for structured client-server interaction

9. πŸ” Access Token Refresh Flow #

  • Automatic renewal of expired access tokens using refresh tokens

10. 🧭 Backend Dockerization #

  • Containerized backend services using Docker

11. πŸ“Š Initial Data Loading on Backend Startup #

  • Automated import of data from files into the database during initialization

12. πŸ“ CI/CD Deployment to Public Server #

  • Continuous integration and delivery pipeline for deployment to a public IP server

13. πŸ—‚οΈ Tutor parser #

  • Automated extraction of tutor information from structured or unstructured sources

14. 🎯 Parallel tutor parsing and MongoDB upload #

  • Simultaneous parsing and storage of tutor data in MongoDB for performance optimization

15. βš™οΈ ML integration with backend #

  • Connecting machine learning modules to the backend for predictions or intelligent features

16. 🌐 User data tracking and server-side storage #

  • Secure storage of user-specific data such as progress trackers on the server

πŸŸ₯ ML #

1. πŸ“Š The code outputs the pros and cons of a suitable profession #

2. πŸ“ Writes the percentage of match #

3. 🎯 Displays a description of the profession, the sphere and the subsphere #

Tech stack #

🟨 Frontend #

Main Framework:
Flutter β€” cross-platform development for Android, iOS, and Windows.

Key Dependencies #

  • percent_indicator: for displaying progress indicators
  • intl: for internationalization and localization
  • pin_code_fields: for PIN code input fields
  • archive and xml: for working with archive files and XML data
  • flutter_secure_storage and shared_preferences: for secure and local data storage
  • cupertino_icons: for iOS-style icons

Development Tools #

  • flutter_test: for unit and widget testing
  • flutter_lints: for static code analysis
  • flutter_launcher_icons: for generating app launcher icons

Data & Resources #

  • Static files:

    • JSON files for universities and professions
    • DOCX files for tests and documents
    • PNG and JPG images for visual assets

  • Localization:
    Handled using the intl package.

🟦 Backend #

Programming Languages #

  • Go
  • Dart
  • Python

Key Libraries and Frameworks #

Go #

  • Web & API: net/http, encoding/json
  • Databases: database/sql (for SQL), mongo-go-driver (MongoDB), github.com/lib/pq (PostgreSQL)
  • Web Scraping: github.com/PuerkitoBio/goquery
  • Security: golang.org/x/crypto/bcrypt
  • Standard Library: context, log, fmt, os, time, errors, regexp, strings, strconv

Python #

  • Web & API: json, fastapi

Databases #

  • SQL

    • PostgreSQL
    • Driver: github.com/lib/pq

  • NoSQL

    • MongoDB
    • Driver: mongo-go-driver

Containerization and Automation #

  • Docker β€” containerization of applications
  • Docker Compose β€” multi-container orchestration and service management
  • CI/CD (GitHub Actions / GitLab CI / Jenkins) β€” automated build, test, and deployment pipelines
  • YAML β€” configuration language for Docker Compose and CI/CD workflows
  • Shell scripting β€” automation scripts within Dockerfiles and CI/CD pipelines

πŸŸ₯ ML #

Programming Language #

Python (3.8)

Used for:

  • Core data processing logic
  • MBTI type determination (determine_mbti)
  • Profession scoring (score_profession_normalized)
  • Recommendation generation (process_student)

Libraries:

  • json: For reading and writing JSON files (e.g., dataset_career_test.json, professions.json, profession_recommendations.json)
  • Potential additional libraries:
    • pandas: For working with structured data (students, professions)
    • numpy: For advanced numeric and statistical operations (e.g., normalization)

Data Format #

JSON

Used for storing:

  • Student data (dataset_career_test.json):
    Includes mbti_scores, subject_scores, interests, values, work_preferences, and exclude.

  • Profession data (professions.json):
    Includes fields like sphere, subsphere, name, description, ege_subjects, mbti_types, interests, values, role, place, style, education_level, salary_range, and growth_prospects.

  • Recommendation results (profession_recommendations.json):
    Includes name, score, positives, negatives, and description.

Advantages:
Lightweight, human-readable format ideal for data exchange between systems and easy local processing.

Database #

File-based storage (JSON)

Current implementation:
All data is stored and managed in JSON files, which is suitable for small-scale projects and local data processing.

Setup instructions #

Clone repository locally:

git clone https://github.com/IU-Capstone-Project-2025/Smartify.git

Starting the app:

cd Smartify/frontend
flutter run

Weekly commitments #

Individual contribution of each participant #

🎨 UX/UI #

Responsible - Kuchukbaeva Regina

  • Privacy policy page
  • Redesign the professions page
  • Page with individual profession

βœ… Related artefacts:

πŸ’» Frontend #

Responsible - Chugaeva Mariia

  • Custom logo display
  • Fixed the main menu page
  • Fixed bugs in the questionnaire field
  • Updated the recommendations page
  • Added the spheres_page.dart page
  • Added the subspheres_page.dart page
  • Redesigned the professions page has been redesigned
  • Updated the design of the cards on all pages
  • The application form button is available from each stage of the profession

βœ… Related artefacts:

Responsible - Zakirov Karim

  • University cards
  • Tracker and calendar
  • The page with tutors
  • Multilingualism
  • Dark theme

βœ… Related artefacts:

πŸ› οΈ Backend #

Responsible - Mayorov Daniil

  • Documentation for all endpoints in the Swagger
  • The ability to save trackers on the server
  • The ability to download trackers from the server
  • Connected everything to flutter so that everything worked on the application
  • Side guides, updated the login logic, now the tokens are still stored on the server before logging in

βœ… Related artefacts:

Responsible - Antipov Alexey

  • Uploading questionnaires to the client
  • Visualize which tables we have in PostgreSQL
  • Visualize mongodb
  • Parsing of tutor profiles

βœ… Related artefacts:

πŸ€– ML #

Responsible - Andruwenko Valery

  • Added 14 areas of work
  • Added sub-spheres for all areas
  • 347 professions now, will be increasing

βœ… Related artefacts:

πŸ“₯ Project Management #

Responsible - Basanov Maxim

  • Establishing communication between each participants in the Telegram
  • Emerging problem solving, meetings organization
  • GitHub repositories, Backlog and all documentation maintaining
  • Demo recording
  • This report writing

βœ… Related artefacts:

Future plans #

πŸ› οΈ UX/UI #

  • Keep the design of the our project beautiful and modern even after the end of the Capstone

πŸ› οΈ Frontend #

  • According to the Figma-design, keep the frontend-part of the our project beautiful and modern even after the end of the Capstone

πŸ› οΈ Backend #

  • Maintain the project and deployment in a working state

πŸ› οΈ ML #

  • Improve MLs metrics
  • Expand all DB

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).