Hello there!
I’m a Data & AI Engineer
ABOUT ME
Education
- INSAT graduate in Computer Science Engineering.
- Currently pursuing an MSc in Data Science at EDC Paris.
Certifications
Google Cloud Professional Data Engineer
Designing data processing systems, ETL/ELT pipelines and BigQuery optimization.
View credential
Google Cloud Professional ML Engineer
Productionizing ML models and deploying scalable ML solutions on Vertex AI.
View credential
Deep Learning / AI Specialization
Comprehensive deep learning concepts: CNNs, RNNs, transfer learning and more.
View credential
Dataiku Core Designer
Validated proficiency building end-to-end data projects in Dataiku DSS.
View credential
What People Say
“Played a key role deploying GenAI agents at scale on GCP.”
Click here to see full endorsement
“Worked extensively with GCP, BigQuery, NLP, and prompt engineering.”
Click here to see full endorsementSERVICES I OFFER
I design practical AI and data solutions that connect modeling, NLP, GenAI, analytics, and cloud deployment into one clear delivery flow.
TensorFlow • Scikit-Learn • Vertex AI
spaCy • NLTK • Hugging Face
LangChain • LlamaIndex • Google ADK
Looker Studio • BigQuery • Streamlit
GCP • Docker • Cloud Run
SKILLS
I focus on mastering many skills and technologies needed for Data & AI Engineering.
EXPERIENCE
Here is a timeline of my key experiences.
- Developed a GenAI labeling agent, automating 80% of workload and reducing labeling costs by 92% (from €500k to €41k).
- Deployed a serverless dual-agent pipeline on Cloud Run Jobs, accelerating batch processing by 50%.
- Reduced LLM token consumption using context caching and batching, reducing input token costs by 90%.
Skills: Python, Google Cloud (Cloud Run Jobs), GenAI/LLMs, prompt engineering, agent orchestration, cost optimization
- Developed Airflow ETL/ELT pipelines to ingest and transform data into BigQuery.
- Created cost efficient data workflows using Airflow automation, reducing BigQuery costs by over 57%.
- Implemented Medallion architecture using dbt for modular models.
Skills: Apache Airflow, BigQuery, dbt, SQL, data modeling, cost optimization
- Deployed GCP Agent Builder chatbot on Cloud Run, automating reporting and batch workflows.
- Built Pub/Sub Dataflow streaming pipelines, reducing latency by 65%.
- Optimized partitioned and clustered BigQuery tables, reducing query costs by 48% across 27M+ rows.
Skills: Google Cloud (Pub/Sub, Dataflow, Cloud Run), BigQuery, streaming pipelines, CI/CD
- Developed a chatbot using Llama 2 LLM (7B parameters) to provide detailed service information to clients.
- Employed RAG techniques to optimize LLM text generation, improving output quality by 30%.
- Integrated and deployed the chatbot application in a web app, managing a CI/CD pipeline.
Skills: Llama 2, RAG (retrieval-augmented generation), vector stores, web deployment, CI/CD
- Developed an NL2SQL system that converts natural language into SQL queries.
- Deployment: Deployed on Google Cloud Run with a CI/CD pipeline via Cloud Build, speeding up processing by 55%.
- Results: Optimized queries and ensured ~95% schema compliance using Dataplex.
Skills: GCP, BigQuery, SQL, Vertex AI, Docker, Git, Prompt Engineering, Tableau
Personal Project
- Implemented a real-time sales analytics pipeline using Google Pub/Sub and Apache Beam, processing 5,000+ transactions/day with under 5s latency.
- Dashboards: Built dashboards in Looker Studio and automated data aggregation into BigQuery.
Skills: Google Pub/Sub, Apache Beam, dbt, BigQuery, Looker Studio, Python
LET'S TALK
If your idea has a pulse, a deadline, or a slightly suspicious spreadsheet, drop it here. I like AI and data projects that are useful, weird, or both.
- No pitch deck required.
- Tea is accepted if coffee is not your thing.
- Short idea? Long idea? I can read both.