Lionel E.

CALF Mission: Lead ML Engineer/Architect (GCP & On-Premise — SAFE)

Objective: Industrialization of ML models and structuring of Cloud/On-prem fundamentals.

Pipelines & Deployment: Design of ML/LLM pipelines (OCR, scoring) on GCP & K8s. On-premise deployment via ArgoCD and Kustomize.

Infrastructure & Security: Network architecture (VPC, Load Balancers), Landing Zone configuration (IAM, AD), and cloud security with DevOps/Security teams.

MLOps Standards: Implementation of CI/CD, versioning, and monitoring (Grafana/FinOps). Creation of templates (Python, UV, Airflow).

Management: Leading LLM industrialization (Capgemini) and monitoring legacy SAS to Cloud migration.

Environment: AWS, GCP, Terraform, Kubernetes, Helm, Vertex AI, Snowflake, Pyspark, Argo, Gitlab.

BforBank Mission: Lead AI Engineer — Agent Platform (Freelance)

Objective: Lead GenAI industrialization and the transition to an intelligent agent architecture.

GenAI Industrialization: Architecture of 6 Fullstack/Backend applications on GKE (200 users). Creation of agentic workflows (Langgraph) integrated with BigQuery, Jira, and Confluence.

Performance & Fullstack: Optimization of React/FastAPI, reducing latency and loading time by 40% (caching, TTL, async). Redesign of the agent management interface.

MLOps & Observability: Definition of K8s patterns (Istio, Flux) and implementation of LLM observability with Langfuse (traces, spans).

Security & Governance: Implementation of guardrails and IAM policies (CISO/DPO). Usage monitoring via Looker dashboards.

Results: Industrialization of 5 key projects (Chatbot, Classifiers) and automation of audits via dbt and K8s Cronjobs.

Environment: GCP (GKE, BigQuery), K8s (Istio, Helm), Python (FastAPI, Langchain), React, Langfuse, dbt.

Objective: Industrialization of global predictive models (50 countries) and standardization of ML/Data platforms on AWS.

Architecture & Scalability: Design of architectures for sales prediction and optimization of massive data volumes on AWS (EMR, S3) and pyspark databricks.

Platform Engineering (MLOps): Standardization of inference and feature engineering pipelines via Databricks (PySpark) and Sagemaker. Implementation of experiment tracking with MLflow and automation via Airflow.

Infrastructure as Code: CI/CD refactoring (GitHub, Jenkins, Terraform) to improve deployment in hybrid ecosystems (AWS, Databricks).

Data Expertise & Monitoring: Design of exposure flows (S3, Kafka) and performance monitoring frameworks via dbt and Tableau dashboards.

Technical Referent & Mentoring: Definition of industrialization standards and support for Data Scientists to raise technical requirements.

Environment: AWS (Sagemaker, EMR, S3), Databricks, PySpark, Airflow, MLflow, dbt, Terraform, Jenkins, Kafka, Tableau, Prophet, DeepAR.

Technical Leadership & Operational Excellence

Talent Development & Mentoring: Coaching and mentoring engineers, technical evaluation (qualification) of profiles for critical missions, and leading specialized training within the internal campus.

Data Architecture & Strategy: Implementation of structuring projects on Data Vault and design of resilient multi-cloud data platforms.

Thought Leadership & Community: Leading the technical community (Tech Cafés), sharing advanced use cases, and evangelizing ML/Data platforms.

Cross-functional Expertise & Problem Solving: Facilitating team communication, resolving complex technical blocking points, and ensuring daily operational fluidity.

Business Vision & Pre-sales: Strategic collaboration with sales teams to define technical scopes, critical requirements, and project roadmaps.

Talent Acquisition: Leading recruitment processes through rigorous evaluation of skills and cultural fit of candidates.