For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
Peter Marinov

Peter Marinov

Lead Machine Learning Engineer - Enterprise AI Systems

United Kingdom flagLondon, United Kingdom
$60.00/hrIntermediateInternal Proprietary Tooling

Key Skills

Software

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code ProgrammingComputer Code Programming
DocumentDocument
ImageImage
TextText

Top Task Types

Classification
Computer Programming Coding
Entity Ner Classification
Evaluation Rating
Segmentation
Text Generation
Text Summarization

Freelancer Overview

I am an experienced machine learning engineer and data scientist with a strong background in designing and implementing data labeling and annotation workflows for NLP, computer vision, and time-series projects. My work spans domains such as insurance, healthcare, e-commerce, and industrial monitoring, where I have led teams in building robust data pipelines, developing entity extraction and classification systems, and optimizing training data quality to improve model accuracy and business outcomes. I am skilled in Python, HuggingFace, TensorFlow, PyTorch, and cloud platforms like Azure and AWS, and have hands-on experience with feedback loops, A/B testing, and production monitoring to ensure continuous improvement of AI training data. My focus is on translating complex data challenges into scalable, reliable solutions that drive measurable results.

IntermediateFrenchEnglishItalian

Labeling Experience

PhD Candidate University of Oxford

Internal Proprietary ToolingImageSegmentation
Summary of the Segmentation Workflow Here’s a step-by-step view of the segmentation and geometry construction approach in the Zacur et al. context: Image Acquisition: Standard clinical cardiac MRI (cine short-axis + long-axis + scout images). Cardiac Structure Segmentation: Expert segmentation of epicardial and endocardial contours across relevant MRI slices. Contour Alignment: Rigid alignment of intersecting slices to correct misalignments due to patient motion or breath holding. Surface Mesh Reconstruction: Algorithms handle sparse, irregular contours to generate 3D surfaces representing heart chambers (ventricles). Torso Geometry Segmentation: Semi-automatic delineation (skin, lungs) from wider field MRI slices; used to define tissue boundaries for body volume conductor modeling. 3D Model Integration: The heart and torso meshes are combined into a personalized anatomical model for computational simulation of cardiac electrophysiology and ECG.

Summary of the Segmentation Workflow Here’s a step-by-step view of the segmentation and geometry construction approach in the Zacur et al. context: Image Acquisition: Standard clinical cardiac MRI (cine short-axis + long-axis + scout images). Cardiac Structure Segmentation: Expert segmentation of epicardial and endocardial contours across relevant MRI slices. Contour Alignment: Rigid alignment of intersecting slices to correct misalignments due to patient motion or breath holding. Surface Mesh Reconstruction: Algorithms handle sparse, irregular contours to generate 3D surfaces representing heart chambers (ventricles). Torso Geometry Segmentation: Semi-automatic delineation (skin, lungs) from wider field MRI slices; used to define tissue boundaries for body volume conductor modeling. 3D Model Integration: The heart and torso meshes are combined into a personalized anatomical model for computational simulation of cardiac electrophysiology and ECG.

2016 - 2021

Education

U

University of Oxford

Master of Science, Computer Science

Master of Science
2016 - 2017
I

Imperial College London

Master of Science, Physics

Master of Science
2011 - 2016

Work History

Z

Zimmer Biomet

Senior Machine Learning Engineer

London
2025 - Present
A

Advantage AI Consulting

Lead Consultant

London
2023 - Present