Available for new projects — Freelance

Turning Environmental Data into Business Decisions

Senior Data Scientist with 7+ years helping research institutes and enterprises build ML pipelines, interactive dashboards, and time-series models that ship. Specialized in climate risk, geospatial analytics, and forecasting.

7+Years of Experience
Ph.D.Marine Ecology
6Peer-Reviewed Papers
2CRAN Packages
Stéphane Karasiewicz — Freelance Senior Data Scientist specializing in climate risk and ML
Trusted by research & industry
📦 12+ ML Projects — Delivered freelance
📊 8,000+ — CRAN Downloads

Data science services that
move your business forward

I don't just build models — I deliver business outcomes. Every project starts with your goals and ends with measurable impact.

Core Expertise
🌍

Environmental & Geospatial Analytics

Satellite data analysis, climate risk modeling, ESG reporting, and environmental monitoring. 2TB+ satellite data management experience.

MODISSentinelNetCDFRisk Assessment
Core Expertise
📈

Time-Series Modeling & Forecasting

Demand forecasting, anomaly detection, and predictive analytics. Full lifecycle from preprocessing to production deployment with LSTM, XGBoost, and Prophet.

XGBoostLSTMProphetscikit-learn
Core Expertise
📊

Interactive Dashboards & BI

Interactive analytics dashboards with country-level maps, KPI tracking, and real-time monitoring. 5+ production dashboards delivered.

DashStreamlitPlotlyShiny
⚙️

MLOps & Data Pipelines

End-to-end ML infrastructure: automated training, CI/CD, Docker containerization, cloud deployment, and ETL pipelines at scale.

DockerCI/CDPostgreSQLREST APIs
🔬

Statistical Consulting & R Packages

Advanced multivariate analysis, time-series modeling, and custom R package development. Published author on CRAN.

PCATime-SeriesNiche ModelingCRAN
🤖

NLP & Computer Vision

Text classification, document extraction, image recognition using state-of-the-art transformer and CNN architectures.

PyTorchTransformersCNNLLMs

Tools & Technologies

PythonRscikit-learnXGBoost TensorFlowPyTorchDashStreamlit PlotlyShinyDockerPostgreSQL NetCDFMODISSentinelAirflow FastAPIGitGitHub ActionsPandas NumPy

From research to production

A track record spanning marine research, freelance consulting, and industry ML engineering.

Present
Freelance Senior Data Scientist
Self-employed — Available for new projects
Short Freelance Missionsfrom 2 to 30 days
Feb 2024 — Jan 2026
Senior Data Scientist
— Data Science Team
  • ML infrastructure for peril risk assessment
  • Engineered end-to-end ML pipeline
  • 3 interactive dashboards with country-level maps
  • Automated data management for 2TB+ satellite data
Jul 2023 — Feb 2024
Freelance Data Scientist
Self-employed — 12+ ML Projects Delivered
  • Customer churn models (XGBoost), price forecasting (LSTM)
  • Full ML lifecycle: preprocessing to deployment
Aug 2022 — Jul 2023
Data Scientist Certification
— 400+ hour program
  • 10 projects: regression, classification, clustering, NLP, CV
  • Deployed 3 ML models via REST APIs with CI/CD
  • Interactive dashboards with Streamlit and Dash
Oct 2021 — Mar 2022
Research Engineer
Boulogne-sur-Mer
  • Automated 500GB+ data extraction — saved 50h/month
  • 70% computation time reduction via script optimization
  • Mentored 4 researchers in R, reproducible research
Sep 2020 — Sep 2021
Postdoctoral Researcher
Boulogne-sur-Mer
  • Bloom detection algorithm on 10+ years satellite data
  • 50,000+ km² coastal analysis (MODIS, Sentinel)
  • Published BDAlgo R package on CRAN
  • 2 peer-reviewed publications, 1 dataset
Sep 2019 — Sep 2020
Freelance Data Scientist
Data analysis & translation
  • Statistical consulting and data visualization for 15+ clients
  • Script optimization: 40-60% performance gains
Feb 2018 — Jul 2019
Postdoctoral Researcher
Centre Brest
  • Analyzed 15+ years satellite data (1TB+)
  • Multivariate techniques (PCA, niche modeling)
  • Mentored 2 Master's students
  • 2 first-author papers, 1 dataset
2014 — 2017
Ph.D. in Marine Ecology
Université de Lille — Phytoplankton & Global Change
  • Novel statistical framework (WitOMI) for niche shifts
  • Analyzed 20+ years monitoring data (10M+ records)
  • 3 first-author papers, 2 CRAN packages (subniche)

Projects with measurable outcomes

Real deliverables, real impact — not just notebooks.

🛰️
Geospatial

Satellite Image Analysis Platform

Deep learning system for land use classification and change detection from satellite imagery using PyTorch with ResNet/EfficientNet and interactive Folium maps.

→ Multi-temporal change detection
📈
Retail / Supply Chain

Time Series Demand Forecasting

Multi-model forecasting combining Prophet, LSTM, and XGBoost for retail demand prediction with seasonality detection, anomaly handling, and ensemble forecasts.

→ Ensemble forecast with backtesting
📊
Business Intelligence

BI Dashboard Suite

Interactive business intelligence dashboard with Plotly Dash and DuckDB featuring KPI cards, cross-filtering, drill-down charts, role-based access, and PDF/CSV export.

→ Multi-page BI dashboard with RBAC
🔍
Finance

Real-time Fraud Detection System

Production-grade ML pipeline for credit card fraud detection with real-time streaming inference, SHAP explainability, A/B testing, and a Streamlit monitoring dashboard.

→ Multi-page real-time dashboard
🔧
Manufacturing / IoT

IoT Anomaly Detection System

Unsupervised anomaly detection for manufacturing IoT sensors using Isolation Forest, PyTorch Autoencoders, and DBSCAN with real-time scoring and alerting.

→ Real-time anomaly scoring pipeline
🔄
Data Warehousing

ETL Orchestration Pipeline

Airflow-based ETL pipeline with multi-source extraction, dbt transformations on DuckDB warehouse, SCD Type 2 snapshots, data quality checks, and lineage tracking.

→ Automated ETL with data quality
Featured Case Study

Climate Risk Data Infrastructure at Raincoat LLC

End-to-end ML pipeline, 3 interactive dashboards, and automated satellite data management for parametric insurance — serving 10+ countries.

Read the full case study →
See All Projects →

Trusted by researchers & engineers

“His coding skills are unparalleled, producing efficient and well-documented codes. His ability to translate our needs is exemplary, and he excels in proposing innovative solutions that go beyond the initial objectives. Stéphane embodies excellence and will be an invaluable asset to any organization.”
Kelig Mahé Head of Unit, Fisheries Resources — Channel & North Sea
“I worked with Stéphane at the beginning of my PhD, and I am still grateful — without his recommendations I would work and troubleshoot much less efficiently. He always takes time to explain, and never loses his sense of humor even after the nth error message.”
Carolin Neven Ph.D. Student
“Steph is a seriously talented engineer with a rare combination of skills. He’s just as comfortable building scientific algorithms and doing deep data analysis as he is shipping polished frontend dashboards — and he does all of it at a high level. What stood out most was how independently he operated: you could hand him a complex project and trust him to run with it, communicate clearly along the way, and deliver something that genuinely moved the needle for the company. I’d recommend him to anyone in a heartbeat.”
Jan Paral Head of R&D and Co-Founder

From question to solution

01

Discovery Call

We discuss your data, goals and constraints. Free, 15 min, no commitment.

02

Scoping & Proposal

Clear approach, timeline, deliverables, and pricing — fixed or time-and-materials.

03

Build & Iterate

Weekly check-ins, shared repos, transparent progress. You're never in the dark.

04

Deliver & Support

Polished deliverables, documentation, and handoff support to your team.

Let's talk about your data

Book a free 15-minute discovery call. No pitch, no pressure — just an honest conversation about how data science can help your business.

Book a Discovery Call → Email Me Download CV
📍 Boulogne-sur-Mer, France 🌐 Working remotely worldwide 🇫🇷🇬🇧 French & English native ⚡ Response < 24h