Bastien LOUVEL

– I help startups, scale-ups and SMEs design, build and scale modern data platforms and AI solutions, with a pragmatic, business-driven approach focused on real impact and operational excellence. – Data & AI Strategy / Advisory – Strategic support for AI adoption (analytics & generative AI) – Data & AI maturity assessments (organization, stack, governance) – Definition of data & AI roadmaps aligned with product and business goals – Delivery of AI training programs and workshops – Identification and framing of high-value AI use cases – Advanced task automation using agentic workflows (LLMs, agents, orchestration) – Selected client: Aleph Studios Data & AI advisory and strategic support – Delivery - Data Engineering & Data Ops – Design and implementation of modern data platforms – Development of scalable and reliable data pipelines – Adoption of DataOps & Analytics Engineering best practices – Production-ready data & AI use cases – Improvements in data quality, reliability and observability – Selected client: Respire Data Engineering & Data Ops missions – Contexts & Industries – Strong background in retail / e-commerce – Product-oriented, fast-growing environments – Data teams scaling from early to mature stages

– Driving Worklib towards a data-driven approach, I design and deploy modern data infrastructure while fostering a data-centric culture across the organization. – I manage a team of four specialists, including a data/analytics engineer, an expert data analyst, a product data scientist, and a growth engineer. – As part of the c-suite, I work with all teams to embed data-driven decision-making processes. – Modern Data Platform: I have implemented a state-of-the-art data platform using leading tools and practices: – ETL: Data integration through Stitch, Airbytes, Airflow, and DBT. – Datalake & DWH: Structuring with Google Cloud Storage (GCS) and BigQuery. – Orchestration & Deployment: Managing Kubernetes (GKE) deployments with Helmchart. – Tracking & Analytics: Advanced solutions like Air360, Google Analytics, GTM, and Segment. – Reverse ETL: Using Hightouch and Airflow for data activation. – Machine Learning: Developing ML platforms with BigQuery, Airflow, and FastAPI. – Data Visualization: Deploying Metabase and Streamlit on Kubernetes. – Infrastructure as Code (IaC): Managing via Pulumi. – CI/CD: Workflow automation with GitHub Actions. – Projects and Achievements: – Cockpits Dashboards: Creation of dashboards to enable the internal team and clients to manage their business. – Recommendation & Search Ranking: Developing engines and algorithms to enhance user experience. – Strategic Analyses: Conducting analyses on pricing, white zones, and user behavior. – Lead Scoring & Automated Prospecting: Implementing systems with AI technologies. – Exploration of LLMs: Promoting Large Language Models with Dust for internal uses like bot creation, Zendesk integration, and SEO optimization.

– Switching the entire data platform to terraform: – S3 buckets, Bucket life cycle policy, EFS, EC2 with airflow celery executor intallation (broker on Redis, backend on RDS) and pentaho, IAM role and policies, DNS... switching the whole stack to private IP. – Optimization of the tracking data archiving stack: – Snowplow and kinesis, Spark with EMR, S3 buckets and Snowpipe to send the data in Snowflake. – Implementation of a CDC solution on the backend database to feed the Snowflake datawarehouse: – Postgres RDS, Debezium, Kafka, Snowflake Sink connector then S3 sink connector, dag to compact the raws tables automatically from the topic avro schema. – management of a person on this project. – Accompaniment of the backend teams in the realization of guidelines on the good practice of production and consumption on Kafka.

– I'm in charge of maintaining and improving the Leboncoin Data Platform. It is hosted on AWS Cloud platform and based on multiple cloud services (S3, Redshift, RDS, Athena and EC2) – I am working with Airflow for ETL orchestration of jobs written in Python, Java. – ETL jobs are mainly implemented using Spark or SQL scripts. – All stack is dockerised and use Ansible and Terraform for infrastructure deployment and the provisioning. – My main responsibilities are : – Integration of front-web data (AT Internet) on the Data Platform, stored in parquet format on S3 Datalake. – Integration of messaging data, creation of KPIs on Redshift facts tables and then creation of the related data source on Tableau. – Integration of Front&Backend data stored in parquet format, partitioned on S3 Datalake and then insert those data into Redshift and provide access to those data through Athena and Spectrum. – Daily and Hourly Data export on other Data Platform and near real-time export with Kafka-Stream. – Support on new Data Platform fork based on the current LBC model, provide training for our tools and development process. – Creation of Data API with Flask RestPlus and DynamoDB on AWS. – Automation of training and deployments of a recommendation model with Sagemaker and Airflow.