Lazar is available for hire

Lazar Stamenkovic

Verified Expert in Engineering

Machine Learning Developer

Location

Niš, Serbia

Toptal Member Since

August 11, 2021

Lazar is a machine learning engineer with four years of professional experience. He is enthusiastic about building AI-based solutions, from development to deployment. He's acquired strong skills in coding and algorithms by participating in various programming competitions and working in the industry. Lazar has experience working in both companies' and universities' research teams, and he always endeavors to contribute with his knowledge and ideas.

Artificial Intelligence (AI)Machine Learning Deep Learning Algorithms Deep Neural Networks Natural Language Processing (NLP)Python Python 3 TensorFlow Keras NumPy Computer Vision FastAPI Web Scraping Data Visualization Natural Language Generation

Portfolio

United Cloud

Machine Learning, Computer Vision, Image Recognition, Video Processing...

Zucchabar Inc.

Artificial Intelligence (AI), GPT, Generative Pre-trained Transformers (GPT)...

Zucchabar Inc.

Algorithms, Natural Language Processing (NLP)...

Experience

Artificial Intelligence (AI) - 4 years Python - 4 years C++ - 4 years Deep Learning - 4 years Machine Learning - 4 years Generative Pre-trained Transformers (GPT) - 3 years Keras - 3 years GPT - 3 years

Availability

Part-time

Preferred Environment

Linux, PyCharm, Slack, Skype, GitHub

The most amazing...

...project I've proposed and developed is a question-answering system that had better performance than the company's previous approach.

Work Experience

Machine Learning Engineer

2020 - 2022

United Cloud

Collected, analyzed, and generated large datasets for image and video processing tasks like channel logo detection and fingerprint recognition.
Developed deep learning solutions for the mentioned tasks that speed up the entire system more than 100 times versus previous company approaches.
Collaborated on a project for human detection and segmentation.
Served production-ready solutions using a FastAPI Python framework, Apache Kafka, the Tensor RT deep learning inference engine, and Docker.
Cooperated on semantic video segmentation and event localization, from research to development.

Technologies: Machine Learning, Computer Vision, Image Recognition, Video Processing, TensorFlow, PyTorch, Keras, FastAPI, Apache Kafka, Docker, Deep Learning, Image Processing, APIs, Convolutional Neural Networks (CNN), Deep Neural Networks, Artificial Intelligence (AI), Image Analysis, Neural Networks

Machine Learning Engineer

2019 - 2021

Zucchabar Inc.

Developed and proposed a new approach for question-answering that outperformed the company's previous one.
Collected the data and developed a model that recommended study groups for users based on their biography and interests.
Worked with other team members on the system for students assessment, including questions' generation from the completed course.

Technologies: Artificial Intelligence (AI), GPT, Generative Pre-trained Transformers (GPT), Natural Language Processing (NLP), TensorFlow, Keras, Flask, MongoDB, Git, Docker, Vue, APIs, FastAPI, Recurrent Neural Networks (RNNs), Convolutional Neural Networks (CNN), Transformers, Text Embedding, Machine Learning, Amazon Web Services (AWS), Amazon S3 (AWS S3), Neural Networks, Recommendation Systems

Junior Machine Learning Engineer

2018 - 2019

Zucchabar Inc.

Developed an algorithm for text classification which assigns the tree of topics to a given text.
Created a project template for deploying machine learning models using Flask and Flask-RESTPlus. The entire deploying flow was unified and speeded up.
Built the annotation and testing tool using Flask, Vue.js, and MongoDB.

Technologies: Algorithms, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), GPT, Python, Python 3, Artificial Intelligence (AI), Machine Learning, Text Classification, Flask, Flask-RESTful, Pandas, Neural Networks, Deep Neural Networks

Junior Software Engineer

2017 - 2018

Freelance

Implemented CRUD API endpoints using Flask and SQLAlchemy.
Created new UI components, integrated the Vuex store, and implemented API calls in Vue.js.
Scraped web resources using Python programming language and libraries such as Scrapy and Selenium.

Technologies: Vue, Flask, Django, SQL

Experience

Codiss.Ai

Codiss.Ai is a web application and chrome extension for saving and semantic searching code snippets. The application consists of the following parts:
• A RESTful FastAPI application.
• A UI Vue.js application.
• A search engine implemented by a vectors' similarity search and text embedding DL model.
• A Chrome extension that enables a more effortless saving of code snippets.

Codiss.Ai is a personal project.

Optimization of a Health-service Schedule

This project aimed to optimize a health service schedule by predicting patients' cancellations and absences. The prediction is based on a model trained with actual data collected in several health institutions in Serbia from 2010 to 2019.

The research paper presented at the ICIST 2020—10th International Conference on Information Society and Technology— at Kopaonik, Serbia, to which I've contributed, is available in the link above.

Herbs Classification

Herbs Classification is a Windows desktop application for classifying images of herbs and getting information about them. The dataset was collected manually and with scraping. Convolutional neural networks were used, such as VGG, ResNet, and Inception Net.

Herbs Classification is a faculty project of mine.

Education

2018 - 2020

Master's Degree in Software Engineering

Faculty of Sciences and Mathematics - Nis, Serbia

2015 - 2018

Bachelor's Degree in Computer Science

Faculty of Sciences and Mathematics - Nis, Serbia

Certifications

SEPTEMBER 2021 - PRESENT

Build a Back-end REST API with Python and Django – Advanced

Udemy

Skills

Libraries/APIs

TensorFlow, Keras, NumPy, Pandas, Scikit-learn, Vue, SpaCy, Vuex, Flask-RESTful, PyTorch, OpenCV, Matplotlib, REST APIs

Tools

Git, Jupyter, PyCharm, Slack, Skype, GitHub, Named-entity Recognition (NER), Docker Compose

Languages

Python, Python 3, C++, JavaScript, SQL, Java, R, C#

Frameworks

Flask, Vuetify, Django, Selenium, Windows Presentation Foundation (WPF)

Platforms

Linux, Docker, Apache Kafka, Amazon Web Services (AWS)

Storage

MySQL, MongoDB, PostgreSQL, Amazon S3 (AWS S3)

Other

Artificial Intelligence (AI), Machine Learning, Deep Learning, Natural Language Processing (NLP), Algorithms, Text Embedding, Logistic Regression, Convolutional Neural Networks (CNN), Deep Neural Networks, Recurrent Neural Networks (RNNs), Transformers, Neural Networks, GPT, Generative Pre-trained Transformers (GPT), Computer Vision, FastAPI, Image Recognition, Video Processing, Natural Language Understanding (NLU), Natural Language Generation (NLG), Search Engines, Image Processing, APIs, Web Scraping, Data Visualization, Supervised Learning, Unsupervised Learning, Text Classification, Sentiment Analysis, Image Analysis, Recommendation Systems, Web Crawlers

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring