Freelancer profile translated to English.

Description

Software, data, and GenAI platform industrialization — polyglot monorepo (Python/TS/Go/Rust) with Bazel, hardened CI/CD (lint, tests, audit, security scan), agents (LangGraph, MCP), RAG (Qdrant), observability and FinOps (Langfuse), multi-provider gateway (LiteLLM), and open-weight model self-hosting (vLLM, SGLang) on GPU Kubernetes. 15+ years of software engineering, ~8 years Cloud Native DevOps (GCP, K8s, Terraform). Your Dev and Data Science teams deliver faster, on standardized environments, with controlled costs and GDPR-compliant data governance. Scoping, build, run.

Industry field of expertise

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Paris (up to 50km)

PwC
Delivery Manager
CONSULTING AND AUDITS
October 2024 - Today (1 year and 8 months)
Paris, France
Supervision of all applications deployed in production on Kubernetes clusters
Coordination of Dev and DevOps teams prior to production releases: configuration review, manifest validation, technical prerequisite alignment
Governance and continuous improvement of CI/CD pipelines: integration of formatting, linting, testing, building, auditing, and security scanning steps
Implementation and governance of a polyglot monorepo (Python, TypeScript/Node, Go, Rust) structured around Git submodules and internal packages
Standardization of builds via Bazel: centralization of dependencies and lock files, execution of tests, production of libraries and binaries
Provisioning of standardized infrastructure and software environments for Lead Developers and Data Scientists for their Proofs of Concept
Implementation and operation of GenAI observability and orchestration tools in self-hosting: Langfuse, Qdrant, Langflow, n8n
Design and deployment of MCP servers in Python, TypeScript, and Go
Consumption of LLMs from major providers (OpenAI, Google Vertex AI, Anthropic, AWS Bedrock, Mistral) via an internal LiteLLM gateway
Deployment of REST applications embedding ReAct agents (LangGraph) with native tools and MCP servers
Implementation of RAG (Retrieval-Augmented Generation) systems: document ingestion, embeddings, semantic search on Qdrant, and integration with agents
Implementation of authentication and access management via OAuth / SSO
Monitoring of LLM inference costs via Langfuse: configuration of pricing per model, aggregation of costs per project and team, analysis of consumption drifts
Deployment and operation of open-weight LLMs in self-hosting
Bazel Kubernetes Python DevOps GenAI
Manutan
Technical Lead
E-COMMERCE
July 2023 - December 2023 (5 months)
Île-de-France, France
Acted as Tech Lead at Manutan on a technical management and steering mission. Next.js stack organized in a Turborepo monorepo, deployed on Kubernetes (Azure). Role focused on the team's daily activities: systematic Pull Request reviews, code quality control, architecture and technology choices aligned with business needs, developer support.
Azure Next.js DevOps Kubernetes Turborepo
Anywr Canada
Full-stack Developer
TECH
September 2022 - April 2023 (7 months)
Montreal, QC, Canada
DevOps and Full-stack Developer mission at Anywr in Canada, combining cloud infrastructure design and application development. DevOps scope: GCP infrastructure design and operation, Ansible automation, GitLab CI/CD pipelines, service containerization. Development scope: Golang backend microservices and Svelte frontend.
Google Cloud Platform (GCP) Ansible Gitlab Kubernetes Golang