Kostiantyn Pavlukhin
Private message
Offer work
Kyiv,
Ukraine
available for hire
Rating
CV
In my collaborations, I prioritize establishing strong partnerships with clients, carefully considering all nuances associated with project execution.
One of my main areas of expertise is programming in the fields of deep machine learning and artificial intelligence.
Examples of my work include:
⭐Natural Language to SQL Queries
⭐Career Buddy - Career Guidance AI Chatbot
⭐Diabetes Prediction Machine Learning Project
⭐Language Translator Application
⭐CodeHub AI Code Generator
Detailed descriptions of completed projects can be found in my portfolio.
Working with a team
I participate in large-scale projects, where my role involves effectively distributing tasks within the team. We work cohesively and productively, achieving high-quality project outcomes. Major companies rely on us to participate in significant projects, and we are always ready to meet their requirements with a high degree of professionalism.
About me
My name is Konstantin Anatolyevich Pavlukhin. I am an experienced developer in deep machine learning and artificial intelligence. Ask me about: Statistics, Machine Learning, Deep Learning, Natural Language Processing, Large Language Models, and Multi-Model Language Models. I am also a Python developer with extensive experience in creating various applications. My skills include developing Telegram bots, processing tabular data using the Pandas library, creating charts and graphs, integrating artificial intelligence into Telegram bots, and developing websites using Django and Flask. I also have experience creating applications with Tkinter and adhere to best practices such as PEP-8. My skills encompass object-oriented programming, multithreading, asynchronous programming, context managers, and other features of Python.
Work Experience
Freelancer, Freelancehunt - from 2022 to present.
Working on independent projects and participating in team projects.
Soft Skills
✳️Experience working in a team and achieving common goals.
✳️Ability to resolve conflicts and experience in application development in a team environment (SDLC).
Hard Skills
✅Python programming language: OOP, SQL, file handling.
✅Development of web interfaces for websites and web applications using HTML, CSS, and JavaScript.
✅Automation and web scraping using the Selenium and Requests libraries.
✅Experience with version control systems Git and platforms GitHub, GitLab.
✅Website development using Django and Flask.
✅Creating applications using the Tkinter library.
✅Working with tabular data using the Pandas and SQL libraries.
✅Adherence to programming principles such as SOLID, Open/Closed, DRY, KISS, YAGNI, and others.
✅Development of Telegram bots using the aiogram and telebot libraries.
✅Experience with the NumPy library for basic computations.
✅Creating artificial intelligence and data processing using the Keras, TensorFlow, and PyTorch libraries.
Skills and abilities
Programming
-
AI & Machine Learning
from 226 USD for project
-
Bot Development
from 226 USD for project
-
Cryptocurrency & Blockchain
from 226 USD for project
-
Data Parsing
from 226 USD for project
-
Databases & SQL
from 113 USD for project
-
Desktop Apps
from 113 USD for project
-
HTML & CSS
from 113 USD for project
-
Python
from 113 USD for project
-
Web Programming
from 226 USD for project
Services
-
Online Stores & E-commerce
from 226 USD for project
-
Payment Systems Integration
from 68 USD for project
Portfolio
- All works 28
- Python 6
- Web Programming 4
- HTML & CSS 8
- Data Parsing 2
- AI & Machine Learning 3
- Bot Development 5
-
500 USD AI & Parsing Program: Restaurant Menus in Real Time
Data Parsing
#LLM #webscraping #AI/ML
Universal Restaurant Menu Scraping System with AI Analytics.
I developed a scalable solution for automated, real-time data collection of restaurant menus worldwide. The system allows users to input any restaurant website link and receive structured information about dishes, prices, descriptions, and images.
… To achieve high accuracy and versatility, a hybrid architecture was implemented:
1. Traditional scraping (Scrapy, BeautifulSoup) for websites with simple structures.
2. AI module (DeepSeek, Gemini) for analyzing and normalizing complex pages, classifying dishes, and extracting data from unstructured content.
3. Browser automation (Playwright) for processing websites with dynamically loaded data.
This project demonstrates the integration of web scraping, machine learning, and systems engineering to create a comprehensive tool for restaurant market analysis.
-
700 USD Automatic Generation Of SFT Data For LLM
AI & Machine Learning
Developed a script for processing a custom text dataset (~400 entries) using a lightweight language model. Built a solution deployable in Google Colab for automatically generating a new CSV dataset in a "question-answer" format based on the original data for subsequent LLM training.
The system is capable of analyzing each new context, generating relevant questions and answers in real time, and efficiently handling large volumes of data.
-
500 USD AI/ML Car Analytics: Data Parsing & Vehicle Ranking System
Data Parsing
#AI #web-программирование #webscraping
The program is designed to scrape data from popular automotive websites, collecting information on make, model, year of manufacture, key complaints, issue descriptions, solutions, and ratings. A script has been developed for automatic daily updates as needed. An AI module was integrated to analyze vehicles and filter out irrelevant data.
… The analysis results are visualized through charts, providing a clear ranking of the top-performing cars.
-
1000 USD CryptoBot
Bot Development
#cryptocurrency #cryptobot
The main purpose of the bot is to analyze current coin pairs from various exchanges. The bot collects information about coin pairs from different exchanges and provides specialized data that helps assess the probability of profitable trading.
… The bot allows users to quickly access analytical information and make informed investment decisions.
-
69 876 USD Diabetes Prediction Machine Learning Project
AI & Machine Learning
This project includes skills in machine learning focused on predicting the likelihood of a person having diabetes. The project features the implementation of various classification models as well as an Artificial Neural Network (ANN) to address the classification task.
Models Implemented
The following models are included in the project:
…
Logistic Regression
Support Vector Classifier (SVC)
Gaussian Naive Bayes Classifier
Random Forest Classifier
Gradient Boosting Classifier
AdaBoost Classifier
Extra Trees Classifier
XGBoost Classifier
LightGBM Classifier (imported as lgb.LGBMClassifier)
-
60 000 USD Natural Language to SQL Queries
AI & Machine Learning
Speak2SQl
This web app is built to convert the natural language to sql queries. It is powered by 2 state of the art Transformer model,
GEMMA 2B
… Deepseek Coder 1.3B
Both the models are fine tuned over 3000 rows and have been tested againest 100 seperate rows. The evaluation metrics used for evaluating these models are ~
BLEU SCORE
SCAR BLEU SCORE
EM RATIO
Here BLEU SCORE considers the ngram probability and checks that if the predicted text matches with the actual text over n-gram, whereas EM Ratio checks that the predicted query exactly matches with the actual query or not.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site
HTML & CSS
This block creates a table and contact details on which you can see the estimates and contact if necessary.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site
HTML & CSS
This block shows a point and a precise description.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site
HTML & CSS
Here is a title and a picture with a demonstration of how it is all done.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site 5
HTML & CSS
In this block, an image is implemented by comparing how it looks if you do repairs and list +.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site
HTML & CSS
This block has a description of the points.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site 3
HTML & CSS
When creating the code, the finished result is displayed. This block shows a scheme and text, as well as icons, and a beautiful background.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site 2
HTML & CSS
This is the visualization of the site. This style.css file created styles for beautiful visualization: text, images, tables.
-
23 USD HTML5 & CSS3 - Creating a web page for an internet site
HTML & CSS
I created a website for the Internet store. In this code there was an insert: text, images, tables. In the index file, the structure of the site was created.
-
21 USD Python – Super Chat 4
Python
After the correspondence is completed, you can press the exit button and the interlocutor will also receive a message about the interlocutor’s exit. After completion, the whole letter is wiped and if you want to recharge you can also press the "recharge" button. And so it got a great application.
Reviews and compliments on completed projects 8
14 August 2025
226 USD
Stage 2, GCP setup
Completed with quality and on time, according to the specifications.
27 July 2025
339 USD
Google Cloud Platform (GCP) Setup
Everything was done according to the technical specifications, on time and with quality.
| Response review
17 May 2025
282 USD
Record a detailed video tutorial on creating an AI agent in Telegram
All right
Thank youʼ
amavmvamivamiva
21 April 2025
282 USD
Парсинг
Thank you, it seems everything has been done well.
| Personal | Response review
3 April 2025
113 USD
Automatic generation of context-oriented SFT data for LLM
The work was done qualitatively, even ahead of schedule. An alternative and more effective solution to the task was proposed.
6 December 2024
113 USD
Data parsing and basic structure services
The first stage of cooperation was completed much earlier than the scheduled deadline! Let's keep working!
17 June 2024
113 USD
Telegram Bot, Cryptocurrency Arbitration Scanner
Expensive, but the project is done.
25 October 2023
113 USD
Development of Telegram-Bot
Very productive interaction. The project is implemented in time. To discuss all the issues on the project. Ready for further cooperation.
