Attila Herbert, Developer in Budapest, Hungary
Attila is available for hire
Hire Attila

Attila Herbert

Verified Expert  in Engineering

Artificial Intelligence Developer

Budapest, Hungary
Toptal Member Since
June 6, 2022

Attila is a Molecular Bionics graduate with a master's in artificial intelligence. His passion for AI extends to deep learning and computer vision. Attila is experienced in the whole production process, from prototyping through development to deployment and optimization. He is comfortable working in diverse teams, keeping up with field advancements, and discussing them with colleagues. Attila is creative and a problem solver who adapts quickly and efficiently, even under pressure.


Deep Learning, Computer Vision, Time Series Analysis, Object Detection...
Python, TensorFlow, Deep Learning, Explainable Artificial Intelligence (XAI)...
Deutsche Post
GraphQL, Neo4j, Python, Scikit-learn, Data Science, Data Engineering, Logistics




Preferred Environment

Visual Studio Code (VS Code), MacOS, PyTorch

The most amazing... I've developed is an explainable AI (XAI) method for semantic segmentation that surpasses state-of-the-art methods.

Work Experience

AI Engineer

2021 - 2023
  • Built a foreign object detector on a conveyor and deployed it on a Jetson edge device.
  • Created a general image-based regression model for tasks on mining sites, ranging from ore analysis to safety warnings.
  • Developed a method for detecting visibility deterioration due to extreme weather or dust accumulation, which is crucial when using computer vision in open-pit mines and similar environments.
  • Trained time series models for autonomous control of production processes.
  • Optimized models for edge computing using TensorRT and other neural network optimization frameworks.
  • Researched and implemented advanced time series prediction techniques, creating a model that outperformed state-of-the-art models in the company's specific use case.
  • Generated LinkedIn content using ChatGPT for PR purposes and to increase company exposure.
Technologies: Deep Learning, Computer Vision, Time Series Analysis, Object Detection, Edge Computing, NVIDIA TensorRT, PyTorch, TensorFlow, Python, Artificial Intelligence (AI), Docker, SQL, Machine Learning, Linux, Azure, Git, Jupyter Notebook, YOLOv5, Time Series, Object-oriented Programming (OOP), Python 3, Convolutional Neural Networks (CNN), Generative Pre-trained Transformers (GPT), Keras

Artificial Intelligence Researcher

2020 - 2021
  • Researched and implemented methods for training computer vision models to be robust in special weather conditions.
  • Examined and executed several explainable AI (XAI) methods for image classification models.
  • Created an internal framework for classifying and organizing XAI methods.
  • Organized production code related to XAI methods to simplify the process of contributing to other team members.
  • Developed a novel explanation method for semantic segmentation models that outperformed cutting-edge explanations.
Technologies: Python, TensorFlow, Deep Learning, Explainable Artificial Intelligence (XAI), Quality Assurance (QA), Machine Learning, Amazon EC2, Agile Software Development, Research, Jupyter Notebook, Python 3, Convolutional Neural Networks (CNN)

Data Scientist

2020 - 2020
Deutsche Post
  • Analyzed large amounts of data to extract useful insights for salespeople and managers.
  • Created a new Graph Database in Neo4j for the data analyzed to achieve better querying and visualization capabilities.
  • Reviewed and optimiZed code written by less experienced colleagues.
Technologies: GraphQL, Neo4j, Python, Scikit-learn, Data Science, Data Engineering, Logistics

Computer Vision Algorithm Developer

2019 - 2019
Verizon Smart Communities
  • Introduced new features for object tracking in the company's embedded camera software.
  • Created a visualization tool for object tracking results and checking improvements.
  • Implemented and optimized code for edge computing in C++.
Technologies: Python, PyTorch, Object Tracking, C++, Docker, Machine Learning, Agile Software Development, Deep Learning, Computer Vision, Linux, Edge Computing, Image Processing, Kalman Filtering, Python 3, Convolutional Neural Networks (CNN)

Explanations for Semantic Segmentation
A thesis project for Maastricht University on explanations for semantic segmentation models.

I developed a new method called Proportionality-driven Occlusion Grid (POG), which outperforms current state-of-the-art explanations. I researched and wrote all the code for my thesis project, which involved learning about semantic segmentation and its challenges.

Safe AI for Autonomous Driving
Several German companies launched this project to share information and create a framework for safer AI techniques in autonomous driving. I was part of this project as a consultant on model robustness, which included researching methods to simulate adverse environments and train models on that simulation and augmenting training data for robustness, implementing these methods, and finally, testing if the resulting models are actually safer.

During this project, I had to work with tight deadlines, give presentations to researchers and high-ranking company officers, and respond to any request regarding model safety in extreme environments.

Foreign Object Detector on Conveyors

This project aims to detect any foreign objects in mined ore (natural rock or sediment) transported on conveyors using cameras placed above the conveyor. I was responsible for any AI-related software in this project. I analyzed the provided data to ensure its quality, developed software that trains an object detector, deployed trained models to Jetson devices using TensorRT, and researched methods for streaming results from the edge device to the mine's monitoring software.

Truck Tracker Application

Designed, prototyped, implemented, and deployed an application that tracks mining trucks in an open-pit mine. The trucks are detected, classified, and tracked. I also created a web app that displays the trucks where the user can select and zoom onto specific trucks.

I created the back end using YOLOv8, TensorRT, and FastAPI and the front end with React and Express.js. I also did the model training, including overseeing the annotation process.

The application was deployed to run in Docker on a Jetson Edge unit.
2019 - 2021

Master's Degree in Artificial Intelligence

Maastricht University - Maastricht, The Netherlands

2015 - 2019

Bachelor's Degree in Molecular Bionics Engineering

Pazmany Peter Catholic University - Budapest, Hungary


PyTorch, TensorFlow, Scikit-learn, Node.js, React, Keras


Jira, Git, Jetson TX2, You Only Look Once (YOLO)


MacOS, Visual Studio Code (VS Code), Docker, Linux, Jupyter Notebook, Amazon EC2, Azure


Python, Python 3, C++, SQL, GraphQL, HTML5, CSS3, TypeScript


Agile Software Development, Data Science, Object-oriented Programming (OOP)




Deep Learning, Neural Networks, Programming, Explainable Artificial Intelligence (XAI), Computer Vision, Time Series Analysis, Object Detection, Semantic Segmentation, Artificial Intelligence (AI), Machine Learning, Convolutional Neural Networks (CNN), Research, Data Engineering, Object Tracking, Quality Assurance (QA), Edge Computing, NVIDIA TensorRT, Image Processing, YOLOv5, Time Series, Medical Imaging, Logistics, Kalman Filtering, OCR, Reinforcement Learning, Generative Pre-trained Transformers (GPT), Torch, APIs, Web App Development

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