Freelancer profile translated to English.

Description

Historically a Data & API Architect with over 20 years of experience in critical and regulated environments (Banking, Insurance, Public Sector, Energy, Retail, Media), I have leveraged these skills to serve Generative AI. I specialize in AI for Enterprise Knowledge. With 3 GRAPH-Augmented RAG implementations (HR, AI for IT & Green Eco Design domains), I bring a new AI paradigm to companies. RAGs based on cosine statistical similarity vectorization cannot understand business concepts, leading to random hallucinations and reduced AI trust. I support AI adoption with a Data & Model Centric vision for surgical, frugal, and predictable AI responses.

Languages

English
Fluent
French
Native or bilingual

Workplace preferences

Can work on-site

Paris (up to 50km)

Customer undisclosed
Enterprise Architect / AI Architect
November 2025 - December 2025 (1 month)
Implementation of an assistant for the administration of the Enterprise Mapping Repository.
Complete automation of TOGAF Repository content administration tasks (functional, application & technical parts)
• Implementation of Model-Centric governance
• Implementation of a unified data model for vectorization and storage
• Automated generation of JSON schemas and SQL DDL
• Automated processing of application architecture files
• Extraction and vectorization of TOGAF artifacts.
• Storage in vectorized and structured database
• Implementation of incremental re-vectorization (AI Continuity)
• Regeneration of Enterprise Architecture diagrams, site plans, application matrices, exchange matrices
• Generation of CSV files for ingestion into the Enterprise Repository
• Python coding

Development of an n8n Community Node
• Synchronization semaphore for parallel AI pipelines
• Publication to • Reuse in TOGAF automation pipelines

Performance Improvement Areas:
• Significant reduction in enterprise repository update time (from several days to a few hours)
• Near-total elimination of manual data entry and consistency errors
• Improved traceability and auditability of architecture artifacts
• Accelerated time-to-insight for IT planning decisions
ESN undisclosed
AI Architect
August 2025 - October 2025 (2 months)
Implementation of a virtual assistant for recruitment and consultant placement functions.
Complete task automation
• CV reception from the website then vectorization in the RAG database
• CV scoring against the job description
• CV regeneration by rephrasing experience and bolding keywords from the job description, conversational agent with the consultant via email for iterative validation of this CV optimization process
• CV generation in the company's graphic charter
• Formulation of a synthetic email to the client highlighting the candidate's strengths and attaching the enhanced CV

Performance Improvement Areas:
• Major reduction in application processing time (from several hours to a few minutes per CV)
• Improved candidate/mission matching rate through automated scoring
• Increased recruitment team productivity (refocusing on value-added tasks)
• Homogenization of the quality of CVs presented to clients
Biogaran
Data Architect in Transition
April 2025 - August 2025 (4 months)
TRANSITION mission for the Enterprise Data Architect position during the employee recruitment period. Mission: Assessment of the existing Data platform.
• Establishment of decision support for Data strategy for the Chief Data Officer:
• Production of all activity resumption deliverables.
Execution of PoCs:
• Resumption of the Medallion architecture
• Resumption of the Data architecture in the SILVER zone
• Automated Data Factory perspective and budget reduction
• Resumption of Enterprise Modeling for instantiation in the Data Platform
• Resumption of the data flow mapping governance
• Implementation of automated diagram generation with scope filtering.
• Evaluation of the data platform and relaunch of the architecture (Medallion/Silver), automated DBT generation, budget reduction via automated Data Factory, resumption of data flow mapping.
• Production of all activity resumption deliverables.

Performance Improvement Areas:
• Substantial reduction in infrastructure costs through Data Factory automation
• Acceleration of data pipeline development cycles via automated DBT generation
• Improved visibility and governance of data flows
• Service continuity ensured during the transition period (zero operational disruption)