Cheick Cisse

Worked on two strategic streams within Servier: the design and rollout of an international Data as a Service platform for affiliates, and the delivery of industrial generative AI solutions at scale.

• Designed and developed LLM pipelines based on RAG, LangChain, LangGraph and ADK for business assistants serving 15,000 active users in production
• Built and maintained a Python FastAPI gateway for LLM orchestration and Vertex AI integration
• Deployed AI services on Cloud Run with load balancing, auto-scaling, secure endpoints, Cloud SQL and PgVector
• Implemented monitoring, prompt traceability and response quality observability with Langfuse
• Industrialized data pipelines with BigQuery, Cloud Functions, Composer, Pub/Sub, dbt, Terraform and reusable CI/CD templates

Stack: Python, SQL, FastAPI, BigQuery, Cloud Run, Cloud Functions, Composer, Pub/Sub, Cloud SQL, PgVector, Weaviate, LangChain, LangGraph, ADK, Vertex AI, Langfuse, dbt, Terraform, Docker, GitLab CI/CD

Worked on several GCP initiatives to industrialize employee, finance and retail data flows in an international context.

• Developed ingestion and transformation pipelines on GCP with BigQuery, Cloud Functions, Composer and Pub/Sub
• Designed scalable data architectures and proof of concepts for business analytics needs
• Industrialized transformation models with Python and SQL in BigQuery
• Implemented unit tests, end-to-end tests and regression testing for production pipelines
• Integrated machine learning use cases with Dataiku and BigQuery ML

Stack: Python, SQL, BigQuery, Cloud Functions, Composer, Pub/Sub, Dataiku, BigQuery ML, Terraform, GitLab CI/CD, GCP

Contributed to a strategic marketing initiative aimed at consolidating multiple data sources into automated executive dashboards for COMEX reporting.

• Developed and industrialized ETL pipelines on Google Cloud Platform
• Built Cloud Functions in Python for process automation
• Designed scoring architectures with BigQuery ML, App Engine and Cloud Storage
• Monitored data and pipelines with Stackdriver and PySpark
• Delivered automated dashboards supporting executive performance steering

Stack: Python, SQL, BigQuery, BigQuery ML, Cloud Functions, Cloud Storage, App Engine, Pub/Sub, PySpark, Stackdriver, GCP

Worked on stock forecasting initiatives, helping centralize and prepare data for predictive analytics and long-term supply planning.

• Designed ingestion and transformation flows on Hadoop and Spark
• Structured data pipelines to feed predictive machine learning models with the Data Science team
• Contributed to large-scale cloud data architecture design

Stack: Hadoop, Spark, Python, SQL, HDFS, HBase

Joined a B2B insurance startup to help build a real-time comparison platform aggregating and enriching market data for professionals.

• Developed web scraping scripts in JavaScript for automated collection
• Stored and indexed data in Elasticsearch for real-time search
• Built APIs and dashboards for end users
• Handled transformation and storage with Laravel, MySQL and MongoDB

Stack: JavaScript, Python, Elasticsearch, MySQL, MongoDB, Laravel, REST API

Worked as a full-stack web developer for a software agency delivering websites and internal applications for multiple clients.

• Built internal applications with PHP, CakePHP, HTML, CSS and JavaScript
• Created and maintained web modules
• Provided PHP and CMS training

Stack: PHP, CakePHP, HTML, CSS, JavaScript