ML Engineer · Paris · GCP Certified
Adrien
Morel
Capabilities
Machine Learning
scikit-learn, XGBoost, LightGBM, PyTorch — feature engineering, cross-validation, hyperparameter tuning (Optuna), model evaluation and performance analysis.
LLMs & Generative AI
RAG pipelines, Hugging Face Transformers, LangChain, prompt engineering, OpenAI and Anthropic APIs.
MLOps & Production ML
End-to-end pipelines, experiment tracking (MLflow), model versioning and registry, batch inference and API serving (FastAPI), model monitoring (Evidently).
Cloud & Infrastructure
GCP — Vertex AI, Cloud Run, BigQuery, Cloud Storage, Cloud Scheduler, Artifact Registry. Docker, CI/CD.
Software Engineering
Python (advanced, typed codebases), FastAPI, Pydantic, REST APIs, Git.
Data Engineering
SQL, PostgreSQL, Snowflake, ETL pipelines, object storage (S3-compatible).
Selected Work
Freelance
09/2024 — present
ML Engineer · Independent Projects
Elec Forecast: end-to-end electricity consumption forecasting system. Time-series pipeline with feature engineering, model training, batch inference and monitoring. Previously: Vélib' Forecast Paris — bike availability forecasting across 1,500+ stations on GCP (Cloud Run, BigQuery, Scheduler) with FastAPI REST API, MLflow and automated retraining.
Career Transition
09/2023 — 09/2024
Professional Development · Military Reserve
Advanced geopolitics training (iReMMO Institute) while serving as reserve officer in French Navy HQ. Deepened mathematical foundations (inferential statistics, measure theory).
→ Ranked 2nd / 180 in Navy staff training (PMS) · Kaggle Top 2% worldwide, 200+ upvotes
SNCF Réseau
10/2022 — 05/2023
Data Scientist · Railway Infrastructure
Automated detection of geometric anomalies across the French railway network using statistical methods and spectral analysis. Designed interpretable indicators and visualization tools to support maintenance decision-making.
→ −50% manual analysis time for a team of 12
Transdev
06/2021 — 09/2022
Data Scientist / Data Developer · Urban Mobility
NLP pipeline for customer feedback analysis (Twitter scraping, sentiment classification). Power BI dashboards for fuel efficiency and operational KPIs. Excel/VBA benchmarking app deployed to 150+ users. Migrated historical datasets to Snowflake, automated reporting via SharePoint and Power Automate.
Stack
Language
Python
ML
scikit-learn, TensorFlow, Keras, XGBoost
Experiment tracking
MLflow, Vertex AI Experiments
Pipelines
Vertex AI Pipelines, Cloud Composer, Cloud Scheduler
Serving
Vertex AI Endpoints, Cloud Run, FastAPI, Docker
GCP
Cloud Storage, Artifact Registry, Pub/Sub, GKE
Data
PostgreSQL, Snowflake, Dataflow
Versioning & IaC
Git, DVC, Terraform, GitHub Actions
Monitoring
Vertex AI Model Monitoring, data drift, performance alerts
Certifications
Professional Machine Learning Engineer
Google Cloud
Associate Cloud Engineer
Google Cloud
Associate Practitioner
scikit-learn
Open Source & Projects
elec-forecast
Forecasting
Electricity consumption forecast model. Time-series prediction pipeline using Python and scikit-learn.
bike-forecast-paris-velib
Production ML
End-to-end bike availability forecasting across 1,500+ Vélib' stations. Vertex AI Pipelines, Cloud Run, BigQuery.
skore
Open Source Contribution
Contributor to skore — the scikit-learn collaborative ML project management library.
Available.