Skip to content
View SuperMBA's full-sized avatar
  • Germany

Block or report SuperMBA

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
SuperMBA/README.md

Hi, I’m Margarita Balandina

Medical Data Scientist | Dentist with German Approbation | MSc Data Science

I combine 13+ years of clinical experience in dentistry with Data Science, Medical AI and healthcare analytics. My focus is the intersection of clinical expertise, biomedical data, machine learning and reproducible research workflows.

Focus Areas

  • Medical AI and healthcare analytics
  • Biomedical data analysis
  • Clinical data quality and validation
  • Machine learning for small and imbalanced biomedical datasets
  • MLOps, reproducible ML pipelines and model deployment

Featured Project

Biomedical Spectra Classification and Data Augmentation

My main research project is a reproducible machine learning pipeline for ATR-FTIR biomedical spectra of saliva and gingival crevicular fluid.

The project covers:

  • supervised baseline vs augmentation experiments
  • leakage-safe evaluation and train-only augmentation
  • PCA-based geometry-first analysis
  • factor–PC association analysis
  • clustering stability checks
  • synthetic data quality control
  • summary reports and figure-ready outputs

Repository: nir-ftir

Medical AI Experience

I participated in the training and validation of Diagnocat, an AI-assisted dental diagnostic system. My work included clinical annotation and validation of dental imaging data, review of label quality and assessment of medical consistency based on dental expertise.

Tech Stack

Python, SQL, pandas, NumPy, scikit-learn, matplotlib, Jupyter, Git, Docker, FastAPI, Airflow, MLflow, DVC, Prometheus, Grafana.

Selected Repositories

Selected Repositories

  • nir-ftir — biomedical spectra classification, train-only augmentation, PCA geometry and synthetic-data QC
  • ml-deployment — FastAPI ML service with Docker, Blue/Green deployment, Nginx and GitHub Actions
  • ml_retrain_pipeline — Airflow DAG for conditional ML retraining, metric-based branching and deployment notification
  • hw4-grafana-alerting — ML service observability with Prometheus, Grafana and SLO-based alerting
  • ml_grpc_service — containerized gRPC service for ML model inference with health check, prediction endpoint and protobuf contract
  • customer-segmentation-sql — SQL-based customer segmentation and transaction analytics

Professional Direction

I am interested in roles at the intersection of data science, clinical expertise and healthcare technology:

  • Healthcare Data Analyst
  • Clinical Data Analyst
  • Medical Data Analyst
  • Data Scientist in Healthcare / MedTech
  • AI Data Quality Specialist
  • Medical AI / Dental AI Specialist Based in Germany. Open to roles in Healthcare Analytics, Clinical Data, Medical AI and MedTech.

Pinned Loading

  1. nir-ftir nir-ftir Public

    Reproducible ML pipeline for ATR-FTIR biomedical spectra: supervised evaluation, train-only augmentation, PCA-based geometry analysis and synthetic-data QC.

    Python

  2. ml-deployment ml-deployment Public

    FastAPI ML service with Docker, Blue/Green deployment, Nginx routing, GitHub Actions and smoke testing.

    Python

  3. ml_retrain_pipeline ml_retrain_pipeline Public

    Airflow DAG for ML model retraining with metric-based branching, conditional deployment and Telegram notification.

    Python

  4. hw4-grafana-alerting hw4-grafana-alerting Public

    ML service observability project with FastAPI, Prometheus, Grafana and SLO-based alerting for latency monitoring.

    Python

  5. ml_grpc_service ml_grpc_service Public

    Containerized gRPC service for ML model inference with health checks, prediction endpoint, Python client and protobuf contract.

    Python

  6. customer-segmentation-sql customer-segmentation-sql Public

    SQL-based customer segmentation project using PostgreSQL queries, joins, aggregations, window functions and transaction analysis.