Switch to English?
Yes
Переключитись на українську?
Так
Переключиться на русскую?
Да
Przełączyć się na polską?
Tak

Freeelancers' best works

1000+
  • 127 USD

    Rent_uabot - Apartment Rental

    Bot development
    Chat-bot parser for Telegram

    The technology:
    by Python 3.9
    The Aiogram
    Thoughts
    The BS4

    The purpose of development:
    Create a chatbot in which the user will be able to create a monitoring of new ads at a price not above the desired, and receive a notification within the first minute of placement on the website olx.ua

    Value for the user:
    The user can create monitoring and first click on a new ad, which in turn guarantees the choice of the best apartment among the available options.
    The possibility of adding the best options to the plugins, and returning to them after a while.

    Deadline of development is 4 days.

    Full functionality :
    [ ] Search for ads by filters
    [ ] Announcement of new announcements
    [ ] Best Advertisements
    [ ] Distribution of your own advertisements
    [ ] List of advertisements
    Tariffs and Online Payments
    [ ] Panel of Management


  • 19 USD

    Shutterstock Bott-Parser

    Bot development
    Chat-bot parser ID and description with Shutterstock

    The technology:
    by Python 3.9
    The Aiogram
    The BS4

    The purpose of development:
    Create a chat job that will be able to send the data about the image it takes from the mail sent to him.


    Term of development is 1 day.


  • 25 USD

    Chat-bot for fixing the working time of employees

    Bot development
    Chat-bot for fixing the working time of employees

    The technology:
    by Python 3.9
    The Aiogram
    Thoughts
    The BS4
    Google Maps API

    The purpose of development:
    Create a chat job in which employees of the company will be able to fix their working hours, with an attachment to the locality.
    In turn, the company’s management could review the working time on all employees, download the data in the Excel table.

    Value for the user:
    The user can create notes about the beginning and end of the working day.
    The chatbot accepts only the signs that are fixed from the company’s office.
    Deadline of development is 2 days.

    Full functionality :
    [ ] Arrival to work
    [ ] The completion of the working day
    [ ] Working hours at the chosen time
    [ ] Download statistics in Excel
    [ ] List of advertisements


  • 1500 USD

    Cashback Service

    Python
    Development of a cache service on Python + Flask with the integration of the CPA partnership program on the API.

    I have developed a fully-backend, and connected to the already ready template with small changes.

    In the project, the technology was used:

    by Python 3.9
    The Flask
    by JavaScript
    by jQuery
    Flask-RESTful (for browsers added)
    The Fire


  • Collecting a large amount of data

    Data parsing
    Screening 934348 search results from the buckeyeinstitute.org resource using your own script on Python. The result is a CSV table with a full set of data.

    #Python #Parsing #Parsing #Scraping #requests #pandas #csv


  • Back-end for 'Mapster' travel service

    Python
    The goal was to create a project that accommodates the complex creation of tours, which can be composed of car rental, accommodation, booking airline tickets, cabs, excursions, transfers on the part of the client. We also created the ability to place all these services on the part of our partners. At the same time, the project contains a global admin panel that covers all of the above functions.

    The most challenging part was to create the architecture of the project, which consists of 9 different services that unite into one complex system. We had to create the most suitable UI library, create custom components and make many identical services in a short period of time. The original goal was that all these 9 services had to be created within 3 months. They all had to be multi-currency, multi-lingual, with complex pricing systems for different types of goods. In essence, all these services are marketplaces, but with a much more complex and larger functionality for adding goods. On ordinary marketplaces, goods are added and realized according to a certain template. In our case, all these goods are very different from each other and cannot be fitted to a certain template.

    For the back-end we chose Django Rest Framework because this is a fairly universal web framework with serialization capabilities that reliably protects user data and conveniently organizes data migration. This framework provides its own authentication and authorization mechanisms such as token authentication, OAuth and JWT. It has excellent documentation and a large, responsive community. It also provides more flexible and powerful tools for managing API access rights.

    #Django
    #Docker
    #Python
    #React_Native
    #Vue.js
    #Web_Application
    #Web_Development


  • Vue.js front-end for 'Mapster' travel service

    JavaScript
    The goal was to create a project that accommodates the complex creation of tours, which can be composed of car rental, accommodation, booking airline tickets, cabs, excursions, transfers on the part of the client. We also created the ability to place all these services on the part of our partners. At the same time, the project contains a global admin panel that covers all of the above functions.

    The most challenging part was to create the architecture of the project, which consists of 9 different services that unite into one complex system. We had to create the most suitable UI library, create custom components and make many identical services in a short period of time. The original goal was that all these 9 services had to be created within 3 months. They all had to be multi-currency, multi-lingual, with complex pricing systems for different types of goods. In essence, all these services are marketplaces, but with a much more complex and larger functionality for adding goods. On ordinary marketplaces, goods are added and realized according to a certain template. In our case, all these goods are very different from each other and cannot be fitted to a certain template.

    In this project, we decided to use Vue.js, as it is a very flexible and rapidly evolving framework that fits almost any project.

    When the project started, the version of the framework was at the peak of its relevance. Vue allows you to develop and build a project much faster than React for example. Vue is very easy to learn and therefore popular, so there should be no problems with project support in the future.

    #Django
    #Docker
    #Python
    #Vuejs
    #UX/UI
    #Web_Development
    #Nuxt.js
    #React_Native
    #figma


  • React Native app for 'Mapster' travel service

    Hybrid mobile apps
    The goal was to create a project that accommodates the complex creation of tours, which can be composed of car rental, accommodation, booking airline tickets, cabs, excursions, transfers on the part of the client. We also created the ability to place all these services on the part of our partners. At the same time, the project contains a global admin panel that covers all of the above functions.

    The most challenging part was to create the architecture of the project, which consists of 9 different services that unite into one complex system. We had to create the most suitable UI library, create custom components and make many identical services in a short period of time. The original goal was that all these 9 services had to be created within 3 months. They all had to be multi-currency, multi-lingual, with complex pricing systems for different types of goods. In essence, all these services are marketplaces, but with a much more complex and larger functionality for adding goods. On ordinary marketplaces, goods are added and realized according to a certain template. In our case, all these goods are very different from each other and cannot be fitted to a certain template.

    Currently, there are 2 most popular frameworks for creating cross-platform applications: React Native and Flutter. There was also a choice to create native applications for each mobile OS separately, using the Kotlin and Swift programming languages, but we decided to abandon this idea in favor of cross-platform development, since such development covers all the needs that arose during the creation of the project. In addition, such development takes much less time than native development, which also corresponded to our plans to create an MVP version of the project in 3 months. Then we had to choose between React Native and Flutter, where we initially chose Flutter and implemented the MVP version using it. We then developed and scaled this version, but the optimization of the application deteriorated quite significantly. We conducted our own research, in which we created an identical application using React Native and received a significant increase in the application loading speed from 4 to 10 times, so we continued to develop the application created using React Native. We also noticed that a React Native application takes up much less storage space compared to Flutter. We managed to reduce the size of the application from approximately 250MB to 20MB.

    #Django
    #Docker
    #Flutter
    #Mobile_App
    #Python
    #React_Native
    #UX/UI
    #Vue.js
    #Web_Development


  • 2500 USD

    FaceID: Face Detection and Identification

    Machine learning
    A Python and C++ library for the detection and recognition of persons.

    Continuous before learning model in the process of recognition.

    The video is available to the top of the portfolio in the resume section!
    https://www.youtube.com/watch?v=P45HTL08IaM

    Face detection 0.04 ~ 0.1 sec/frame
    Learning features of each face ~0.2 sec/face on avr CPU
    Face Recognition:
    High accuracy more then 97.5%, Sensitivity ~ 75%
    Very fast searching in faces database (500k faces per second)
    Live learning during the recognition process


  • 2000 USD

    Anonymous Chatbot

    Python
    Anonymous Chatbot
    The book is written in Python, Libya Aiogram.The data base is aiomysql.It holds any burden and asset.(all depends on your server)

    1 .Anonymous chat between users
    2ndSearching for an interlocutor on the floor
    ThreeOpportunity to receive anonymous letters using a personal link
    4 .User profile with statistics

    The admin part (Admin inside the bot)
    1 .Access to the bot by mandatory subscription to the channel/bot (Add and edit)
    2ndDetailed statistics of each link from the advertisement (Call-in the user, how many have passed the registration in the bot, signed on the channels, stopped the bot)
    ThreeThe links from the OP and Messages are processed separately (Required byte text is issued)
    4 .Bot receives applications to channels (Dop.Transfer of users from the channels
    and 5.Detailed statistics of BOT.Show all the activity in the bot (New users in an hour-day, how much the bot has been used in the last 24 hours, how much the bot has stopped, chat, channel, etc.)6 .Advertising posters .Advertising posts for all users and unique (See in different locations).Advertising post for everyone is displayed at the end of the chat 1 time in 6 hours, for unique before registration + stake on posts
    7 .Message in the bot.The speed of sending is as fast as possible as it allows the limitation of the telegram.Notification of the result of the mail + notification to the personal channel (The mail is launched, the mail and who has launched.It is convenient if the ads are bigger than 1, that no one will launch the left ad) + post the start of the mail on the right day/time.8 .Logging of errors.The error is logged in the file, the bot does not fall and continues its work.Errors are always visible in the adminke.The 9.Download BackUp'a user base directly inside the bot
    The 10.Search for locked users

    DOP Functional
    1 .All types of data are supported inside the chat (Pictures, Stickers, Voice, etc.)2ndBoot Registration for New Users
    ThreeThe script checks every sent message for the availability of contact details, links, emails, user reference.Protect from spam.4 .Return of the interlocutor.After the dialogue is completed, the current interlocutor can be asked to start the dialogue again that he must accept or reject.and 5.The boat has a dynamic feedback system.After the dialogue is completed, there is an opportunity to leave a comment on the interlocutor.If a user collects 10 negative reviews over a short period of time, the spamming button will be available to his interlocutors after the dialogue is completed.If not a goodwill user further intensely gathers reviews about himself as a spam, he will be locked in the bot for a certain period of time.After the bath time expires, he will be available to chat.The user reputation is updated periodically.Each user per day is given '10 votes' - the opportunity to leave a comment on the interlocutor 10 times.6 .Before starting the search, a simple drop is made to protect against bots (new drop solution, every 6 hours)