Week6

Practicum Project Summer 2025 Report 6 #

Team mealix

2. Final deliverables #

Project overview

  • Mealix is a web and Telegram-based service for personalized meal planning.

Features

  • Lets users generate daily meal plans tailored to their preferences, allergies, budget, and nutrition goals (calories, proteins, fats, carbs).
  • Users register and log in via Telegram bot (no separate signup needed).
  • Users can save their food profile (allergies, restrictions, favorite cuisines).
  • The frontend provides a form to generate a meal plan for one day, with fields for preferences, budget, and nutrition goals.
  • The backend stores user profiles, checks registration, and generates meal plans by integrating with an ML service.
  • All components (frontend, backend, ML, bot, database) are containerized and run together via Docker Compose.

Tech stack

Frontend:

  • React
  • TypeScript
  • Vite
  • React Router

Backend:

  • Java 21
  • Spring Boot
  • Spring Web, Spring Validation, Spring Data JPA, Spring WebFlux
  • PostgreSQL
  • Liquibase (database migrations)
  • Maven

ML Service:

  • Pytorch
  • FastAPI

Telegram Bot:

  • python-telegram-bot (asyncio)
  • httpx
  • python-dotenv

3. Presentation draft #

Link for presentation draft

4. Weekly commitments #

Individual contributions #

Alexander Kachmazov (ML):

  • Implemented structured response generation based on a predefined JSON schema.
  • Added retry logic for error handling to improve robustness.
  • Retrained the RAG (Retrieval-Augmented Generation) model for improved performance and accuracy.
  • Link

Samat Iakupov (Frontend):

  • Adapted frontend with new backend contract.
  • Link

Dmitrii Antipov (Backend):

  • Adapted backend with new ML contract.
  • Link

Danil Eramasov (ML):

  • Improved recepies deleting all trash from text. Read all recipes in 5 categories.
  • Code link

Albert Shammasov (Data Analyst):

  • Improved recepies deleting all trash from text. Read all recipes in 5 categories.
  • Code link

3. Confirmation of the code’s operability #

We confirm that the code in the main branch:

  • Is in working condition
  • Can be run via docker-compose (or alternative method described in README.md)