Vivien B.

Key Skills

Software

Don't disclose

Top Subject Matter

Biomedical single-cell image analysis (hiPSC-derived cardiomyocytes/cell biology)

Cell biology morphology-based phenotypic classification

Biomedical/Cell biology scientific evaluation (STEM reviewer)

Top Data Types

Image

Text

Video

Top Task Types

Prompt + Response Writing (SFT)

Question Answering

Relationship

Segmentation

Classification

Freelancer Overview

PhD Researcher (CIFRE Industrial Program) — AI-ready dataset production via image annotation pipeline. Brings 8+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose and scikit-learn. Education includes Doctor of Philosophy, Sorbonne University (2024). AI-training focus includes data types such as Image and Text and labeling workflows including Segmentation, Classification, and Evaluation.

Labeling Experience

Technology Transfer Project Manager — Scientific fact-checking and expert review (FR/EN STEM reviewer)

Don't discloseText

Performed scientific fact-checking and structured evaluation of technical and R&D materials in English and French. The role involved assessing the quality and feasibility of innovation projects using domain knowledge and checklist-style technical review. This constituted expert review of AI-adjacent scientific outputs and prepared decisions for strategic next steps. • Conducted TRL assessment and feasibility evaluation for 3D cell culture projects • Performed structured analysis and synthesis of complex R&D datasets • Supported patent co-drafting and pre-industrialization roadmap development • Delivered multilingual (FR/EN) technical scientific judgments for stakeholders

2024 - Present

PhD Researcher (CIFRE) — Morphological feature labeling for ML phenotypic classification

ImageClassification

Defined and validated morphological feature sets used to label and classify biomedical phenotypes for ML tasks. The labeling effort translated measurable image-derived characteristics into structured inputs for model training and evaluation. This supported disease phenotype classification and downstream scientific interpretation. • Labeled morphology-derived features such as size, texture, shape, intensity, and radial distribution • Validated feature sets for classification suitability • Supported ML disease phenotype classification through feature selection • Evaluated and interpreted model misclassifications for annotation and QA feedback

2019 - 2024

PhD Researcher (CIFRE Industrial Program) — AI-ready dataset production via image annotation pipeline

Don't discloseImageSegmentation

Developed and validated a high-content single-cell image analysis pipeline to generate AI-ready annotated image data. The work included segmentation and preparation of labeled crops and masks used for downstream ML training and evaluation. Outputs were shaped for reliable dataset production in a biomedical imaging context. • Performed segmentation-driven mask generation for single-cell images • Produced crop labeling for image-ready training samples • Validated dataset quality for AI annotation readiness • Coordinated labeling pipeline development with ML collaborators

2019 - 2024

Education

Work History