Skip to content
View Adrien-1997's full-sized avatar
0 followers · 5 following

Achievements

Achievement: YOLOAchievement: Pull Shark

Achievements

Achievement: YOLOAchievement: Pull Shark

Block or report Adrien-1997

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
Adrien-1997/README.md

Adrien Morel — Data Scientist · Applied Machine Learning

Paris · Remote-friendly
GitHub · LinkedIn · Email

About

I build practical, data-driven products with a strong focus on clarity, metrics, and reproducibility.

My work usually follows a simple but rigorous path:

Python / SQL → feature engineering → modeling → evaluation → clean exports & documentation

I care as much about what a model does as about how it is built, explained, and maintained.
All my code and technical documentation are published openly on GitHub.

Tech stack

Core — Data Science

Python SQL Pandas NumPy Scikit--learn Statsmodels

Applied Machine Learning

Feature_Engineering Supervised_Learning Unsupervised_Learning Time_Series_Modeling Anomaly_Detection Model_Evaluation

Data & Analytics

Power_BI Streamlit Jupyter Snowflake Parquet

Backend & Product

FastAPI REST_API JavaScript Node.js Next.js

Engineering, CI/CD & Cloud

Git Docker GitHub_Actions Google_Cloud

Selected projects

Bike availability forecasting — Paris (T+1h)

End-to-end forecasting pipeline for Vélib’ bike availability.

  • Feature engineering (temporal, usage-based)
  • Supervised ML forecasting
  • Clear evaluation metrics
  • Public documentation & demo site

Repository
https://github.com/Adrien-1997/bike-forecast-paris-velib

Crime & Safety Dashboard — France (2016–2024)

Exploratory and analytical dashboard focused on readable public indicators:

  • Rates per 1,000 inhabitants
  • Temporal trends & anomalies
  • Emphasis on statistical clarity and storytelling

Repository
https://github.com/Adrien-1997/crime-safety-dashboard-fr

Kaggle — structured learning & clean baselines

Collection of notebooks focused on:

  • Solid EDA
  • Reproducible baselines
  • Clear assumptions and validation

Repository
https://github.com/Adrien-1997/kaggle-learning

How I work

  • Start from a concrete question or metric
  • Build a useful prototype quickly
  • Iterate using real feedback
  • Reinforce with validation, documentation, and light monitoring

Currently open to

  • Data Scientist / Applied ML roles
  • Freelance & collaboration opportunities
  • Projects where data is expected to support real decisions

Popular repositories Loading

  1. crime-safety-dashboard-fr crime-safety-dashboard-fr Public

    Streamlit dashboard for France crime & safety: KPIs per 1,000 from police/gendarmerie data; spatio-temporal views, anomaly detection, baseline forecasts.

    Python 1

  2. Adrien-1997 Adrien-1997 Public

    Config files for my GitHub profile.

  3. kaggle-learning kaggle-learning Public

    Kaggle notebooks organized by competition with metrics, validation, and takeaways — learning in public.

    Jupyter Notebook

  4. applied-ml-playbook applied-ml-playbook Public

    Opinionated notebooks for EDA, baselines, anomaly detection, time series and evaluation — statistics first, reproducible.

    Jupyter Notebook

  5. bike-forecast-paris-velib bike-forecast-paris-velib Public

    Bike-sharing demand forecasting in Paris (Vélib’). A data science and machine learning project leveraging open urban mobility data to predict bike availability, analyze time series usage patterns, …

    Jupyter Notebook

  6. pjm-load-forecasting-benchmark pjm-load-forecasting-benchmark Public

    A unified benchmark of statistical and machine learning models for 24-hour PJM load forecasting.