Petar is available for hire
Hire PetarPetar Sekulić
Verified Expert in Engineering
Machine Learning Developer
Location
Podgorica, Montenegro
Toptal Member Since
June 30, 2021
Petar is a machine learning engineer interested in new science trends and very curious about novelties in this field. He enjoys working on various machine learning projects and expanding his competencies and skills. Petar focuses on learning and experimenting with ML every day.
Portfolio
Toptal Client
Data Science, Machine Learning, Natural Language Processing (NLP), GPT...
Fleka
Python 3, TensorFlow, Elasticsearch, Generative Pre-trained Transformers (GPT)...
Omnitech MNE
Python 3, TensorFlow, Data Science, Machine Learning, Deep Neural Networks...
Experience
Availability
Part-time
Preferred Environment
Python 3, TensorFlow, NumPy, Jupyter Notebook
The most amazing...
...models I've developed were for scanned document processing and provided image to text conversion and semantic analysis of the text.
Work Experience
Data Scientist
2021 - 2023
Toptal Client
- Acted as the director of data science while working on the Document AI project for a global real estate services company.
- Built a sourcing strategy for different document types.
- Worked on a model selection for different parts of the machine learning pipeline.
- Fine-tuned state-of-the-art models for the Document AI project and established quality metrics.
Technologies: Data Science, Machine Learning, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), GPT, Named-entity Recognition (NER), Azure, Google Cloud Platform (GCP), Python, SQL, Data Analysis, Data Reporting, Data Visualization, Data Modeling, Data Engineering, Statistical Analysis
Machine Learning Scientist
2020 - 2021
Fleka
- Worked on a machine learning algorithm to create a smart image database to help journalists make better use of the photo archive.
- Implemented Elasticsearch for the official website of the Government of Montenegro. The site has a large number of documents and articles that need to be processed and placed in Elasticsearch to improve search results.
- Created an algorithm for smart image cropping to automate image cropping for news portals.
Technologies: Python 3, TensorFlow, Elasticsearch, GPT, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), Image Processing, Machine Learning, Data Science, Python, SQL, Data Reporting, Data Visualization, Data Analysis, Data Modeling, Data Engineering, Statistical Analysis
Machine Learning Engineer
2018 - 2020
Omnitech MNE
- Worked on a machine learning system on the SeVaRA project. SeVaRA integrates an R&D project aimed at defining an innovative system for the calculation of an aggregated environmental risk index.
- Wrote a scientific paper, "A Recurrent Neural Network Approach to Improve the Air Quality Index Prediction."
- Organized an eight-week-long machine learning course. Topic: Introduction to Deep Learning and TensorFlow.
Technologies: Python 3, TensorFlow, Data Science, Machine Learning, Deep Neural Networks, Recurrent Neural Networks (RNNs), Python, SQL, Data Reporting, Data Visualization, Data Analysis, Data Modeling, Data Engineering, Statistical Analysis, Generative Design
Machine Learning Engineer
2017 - 2018
Datum Solutions
- Developed a machine learning system for signature recognition. I used deep learning techniques to solve this task.
- Created deep learning algorithms for optical character recognition (OCR) and intelligent word recognition (IWR).
- Worked on an algorithm for semantic text analysis to perform document classification.
- Tracked tasks and bugs using Jira as a reporting tool.
Technologies: Data Science, Machine Learning, Python 3, TensorFlow, Python, SQL, Data Visualization, Data Modeling, Data Engineering, Statistical Analysis, Generative Design
Machine Learning Researcher
2015 - 2017
BIO-ICT Centre of Excellence, Faculty of Electrical Engineering
- Worked on retinal blood vessel segmentation using machine learning techniques.
- Used machine learning and image processing algorithms for the detection of downy mildew in grapevine leaves.
- Wrote three scientific papers explaining which algorithms were used during the retinal blood vessel segmentation and detection of downy mildew in grapevine leaves projects.
Technologies: Python 3, MATLAB, Mathematics, Data Science, Machine Learning, Python, SQL, Data Visualization, Data Modeling, Statistical Analysis, Generative Design
Experience
SeVaRa
https://www.sevara.it/SeVaRA integrates an R&D project aimed at defining an innovative system for the calculation of an aggregated environmental risk index derived from multiple parameters summarized in risk of consequences of very short-term weather events, risk of the geo-location according to historical data, the risk associated with instability of soil or manufactured goods.
My role as a machine learning engineer was to create models that would predict natural disasters such as rain showers, soil erosion, and floods.
My role as a machine learning engineer was to create models that would predict natural disasters such as rain showers, soil erosion, and floods.
RapidCAPTURE
RapidCAPTURE uses AI to leverage natural language processing and machine learning for advanced and cost-effective imaging and document ingestion, converting unstructured content to structured format.
I worked on image-to-text conversion models as well as image processing with the goal of noise removal from the images.
I worked on image-to-text conversion models as well as image processing with the goal of noise removal from the images.
BIO-ICT
https://www.linkedin.com/company/bio-ictBIO-ICT Centre of Excellence is the 1st center of excellence in Montenegro. It is implemented as a three-year research program at the University of Montenegro, led by the Faculty of Electrical Engineering.
I worked as a young researcher at the BIO-ICT Center of Excellence on various machine learning projects. Some of the projects were retinal blood vessel segmentation and the detection of downy mildew in grapevine leaves. This was also part of the research during my master's studies.
I worked as a young researcher at the BIO-ICT Center of Excellence on various machine learning projects. Some of the projects were retinal blood vessel segmentation and the detection of downy mildew in grapevine leaves. This was also part of the research during my master's studies.
Skills
Languages
Python 3, Python, SQL, Java, C, C++
Libraries/APIs
TensorFlow, NumPy, SpaCy
Paradigms
Data Science
Other
Machine Learning, Artificial Intelligence (AI), Data Modeling, Statistical Analysis, Mathematics, Signal Processing, Image Processing, Data Analysis, Data Visualization, Data Engineering, Generative Design, Programming, Expert Systems, Networks, Probability Theory, Electronics, Video Processing, Recurrent Neural Networks (RNNs), Deep Neural Networks, Natural Language Processing (NLP), Data Reporting, GPT, Generative Pre-trained Transformers (GPT)
Platforms
Jupyter Notebook, Azure, Google Cloud Platform (GCP)
Storage
Databases, Elasticsearch
Tools
MATLAB, Named-entity Recognition (NER)
Education
2015 - 2017
Master's Degree in Computer Science
University of Montenegro - Podgorica, Montenegro
2014 - 2015
Specialist Degree in Computer Science
University of Montenegro - Podgorica, Montenegro
2011 - 2014
Bachelor's Degree in Electrical Engineering
University of Montenegro - Podgorica, Montenegro
Certifications
NOVEMBER 2016 - PRESENT
Fundamentals of Digital Image and Video Processing
Coursera
SEPTEMBER 2016 - PRESENT
Machine Learning
Coursera
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.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
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