Open to opportunities

risachan User

@risachanuser

Computer Vision Scientist applying machine learning to real-world, medical problems and delivering empathetic, communication-driven data science.

Germany

Message

What I'm looking for

I’m looking to build impactful medical ML—especially computer vision—working with researchers and clinicians. I want to own models end-to-end, iterate with domain experts, and ship production-ready pipelines while maintaining strong communication and empathy.

I’m a Computer Vision Scientist with 5+ years in applying Machine Learning to real-world problems, with a strong pull toward the medical field because I want to make a difference with my work. My data science is “tech-flavoured” by default, but I also bring excellent communications skills and empathy into cross-functional collaboration.

At Aignostics, I lead and co-lead CNN and Foundational Vision Transformer tissue segmentation on high-resolution medical images (WSIs), building blur detection and quality-control models to filter out artifacts and generate automated QC reports for big cohorts. I’ve worked closely with pathologists to define annotation styles, improve model performance in an iterative bundle, and operationalize pipelines with Prefect/Ray, while at the same time translating insights into practical outcomes for product and client teams.

Experience

Work history, roles, and key accomplishments

Current

Lead Data Scientist

Current

Aignostics GmbH

Nov 2022 - Present (3 years 6 months)

Led and co-led model development for high-resolution medical image (WSI) tissue segmentation, building blur detection and automated QC filtering for large cohorts. Improved model performance in close collaboration with pathologists and implemented distributed inference using Prefect/Ray with experiment tracking in Weights & Biases.

Computer Vision Whole Slide Image (WSI) Segmentation Prefect Ray Weights Biases Docker

Data Scientist

Glooko GmbH

May 2022 - Aug 2022 (3 months)

Conducted code reviews, EDA, data cleaning, and imputation for high-dimensional medical time-series data, delivering product insights and defining glucose control metrics. Built a diabetes classifier that improved performance by 18% and wrote/optimized Snowflake SQL to achieve 10x faster data loading.

Python Jupyter EDA Data Cleaning Imputation Time Series SQL Data Loading Metric Design

Data Scientist

Fev Norddeutschland GmbH

Mar 2022 - May 2022 (2 months)

Developed a Gaussian-process automotive speed prediction model and benchmarked it against LSTM and Markov chain approaches, improving performance to an MRSE of 0.17 m/s. Evaluated explainable ML boundary conditions for a client and delivered a handover roadmap to support further collaboration.

Speed Prediction Benchmarking Model Evaluation Python Time Series Modeling

Data Scientist

Fraunhofer Iais

Apr 2020 - Feb 2022 (1 year 10 months)

Published three papers on kernel methods for numerical optimization, hyperparameter search, and prototype extraction. Developed anomaly detection models for image and keystroke data, improving recall/precision/F2/Gm metrics up to 98% and generating new image datasets using GANs and GENs.

Hyperparameter Tuning Anomaly Detection Keystroke Analytics Image Generation Deep Learning Model Evaluation

Visiting Researcher

FernUni & Max Planck Institute

Mar 2020 - Mar 2020 (0 months)

Conducted reproducibility research using visualization analysis and orchestrated interactive eye-tracking experiments in Python. Handled data acquisition, analysis, and feature engineering to support experimental outcomes.

Eye Tracking Python Data Visualization Data Acquisition Data Analysis Feature Engineering Experimental Design

Education

Degrees, certifications, and relevant coursework

University of Bonn

Master degree, Computer Science (Computer Graphics & Intelligent Systems)

2018 - 2022

Master’s degree at the University of Bonn with a focus on Computer Science, Computer Graphics & Intelligent Systems. Thesis: Practical Kernel Minimum Enclosing Balls.