Skip to content
View anastev982's full-sized avatar

Block or report anastev982

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
anastev982/README.md

Hi, I'm Ana πŸ‘‹

Data Scientist & Analyst Β· Clinical & Scientific Data Β· ML Engineering Β· LLM Integration

13+ years in regulated clinical and scientific environments β€” now building intelligent data pipelines, ML models, and LLM-powered tools.


🧬 About Me

  • πŸŽ“ M.Sc. Data Science candidate @ Singidunum University
  • πŸ₯ Background in nursing + analytical chemistry β€” I understand the data and the domain
  • πŸ€– Passionate about applying ML and LLMs to real clinical and scientific problems
  • 🌍 Based in Serbia Β· Open to remote EEA/EU Β· Available for roles in πŸ‡©πŸ‡ͺ πŸ‡³πŸ‡΄ πŸ‡¦πŸ‡Ή
  • πŸ’¬ Languages: Serbian Β· Norwegian Β· English

πŸ› οΈ Tech Stack

Languages & Data

Python SQL Pandas NumPy Dask

Machine Learning & AI

scikit-learn OpenAI LLM

Tools & Infrastructure

Docker Git Jupyter PyArrow


πŸš€ Featured Projects

Project Description Stack
🧠 LLM Clinical Decision Support NLP pipeline using GPT-4o for clinical decision support with zero-shot classification Python · GPT-4o · NLP
πŸ’Š Anti-Cancer Drug Effectiveness ML pipeline on GDSC biomedical data β€” RF & Gradient Boosting models, large-scale with Dask Python Β· Dask Β· scikit-learn
πŸ“° Fake News: ML vs LLM Comparative study of traditional ML vs GPT-4o on fake news detection Python Β· GPT-4o Β· NLP
πŸ“Š Jobs-in-Demand EDA EDA on 100k+ LinkedIn job postings β€” skill demand trends across data roles Python Β· Pandas Β· Visualization
πŸ”¬ IR Spectra Clustering PCA + KMeans clustering on infrared spectra data β€” domain expertise meets ML Python Β· scikit-learn Β· Chemistry
πŸ€– AI Ops Automation Automated operations framework integrating LLM pipelines Python Β· LLM Β· Automation

πŸ“ˆ GitHub Stats


πŸ“¬ Let's Connect

LinkedIn Email


"Data is most powerful when it comes with domain understanding."

Pinned Loading

  1. llm-clinical-decision-pipeline llm-clinical-decision-pipeline Public

    Policy-guided LLM safety pipeline for medication-related clinical decision support and risk-aware decision routing.

    Python

  2. anti-cancer-drug-effectiveness anti-cancer-drug-effectiveness Public

    From raw biomedical data to predictive modeling: a machine learning pipeline for anti-cancer drug response prediction.

    Python

  3. jobs-in-demand-2025-26 jobs-in-demand-2025-26 Public

    EDA of 100k+ LinkedIn job postings: role mapping, skill extraction, cloud tech trends, and job-market insights.

    Jupyter Notebook

  4. spectromind-ir-spectra spectromind-ir-spectra Public

    PCA & KMeans clustering analysis of IR spectra (JCAMP-DX format)

    Jupyter Notebook

  5. fake-real-news-ml-vs-llm fake-real-news-ml-vs-llm Public

    Fake news classification using a classical Machine Learning baseline (TF-IDF + Logistic Regression) compared to modern LLM zero-shot models (GPT-4o-mini, GPT-4.1-mini, GPT-4o). Includes analysis, e…

    Jupyter Notebook

  6. covid19-analysis covid19-analysis Public

    Time-series analysis of COVID-19 trends for Serbia, Norway and Germany using BigQuery SQL and Python (OWID data)

    Jupyter Notebook