You're seeing this page as if you were . The main menu is still yours, though. Exit from immersion
Guillaume RivalanGR

Guillaume Rivalan

Python Developer | Django | AI Agents, RAG, MCP

€650/day
1 project
Lyon, FR
3-7 years

Average response time: 1 hour

Freelancer profile translated to English.
Back to original language

About Guillaume

Senior Python Developer— Django,Generative AI(AI agents, RAG, LLM, MCP), data engineering, and AWS cloud. Looking for an expert for a robust backend or to integrateartificial intelligenceinto your product? You've found the ideal person!

What I do for my clients:
-Create AI agentsand conversational assistants:GPT-4oagent withRAG(partitionedQdrantvector database), businesstools**, and **MCP(Model Context Protocol) server, integrated into a Django backend.
-Design development frameworksfrom scratch: currently for SNCF, a complete Python framework (architecture, DSL, CLI, AWS runtime) for business teams.
- **Python backend architecture**: Robust and scalable REST APIs and applications (Django REST Framework, hexagonal architecture, AWS serverless).
-High-performance data pipelines(Polars, PostgreSQL, AWS) on large volumes.
- **Full-stack Django / Angular web applications**, from A to Z.

My areas of expertise:
- **Generative AI**: LLM agents (PydanticAI, OpenRouter, GPT-4o), RAG and vector search (Qdrant), MCP creation, integration of AI assistants into existing products.
- **Python backend architecture**: Hexagonal architecture, REST APIs with Django REST Framework, advanced ORM, Pydantic, industrialized quality (uv, ruff, static typing, CI/CD).
- **Data engineering**: High-performance pipelines with Polars, large volumes, advanced PostgreSQL.
- **AWS Cloud & Observability**: Serverless (Lambda, SQS, S3), Datadog monitoring.
- **Angular Frontend**: Material UI / Tailwind, mapping (Leaflet, MapLibre), PowerBI.
- **Analysis & Support**: Specifications, documentation (Confluence), agile (Jira).

I usually respond within an hour. Let's talk about your project!
  • French

    Native or bilingual

  • English

    Conversational

Remote only
Primarily works remotely

Experience

  • SNCF
    Design and development of a railway diagnostic framework - Senior Python Consultant
    LOGISTICS AND SUPPLY CHAIN
    January 2026 - Today (7 months)
    Lyon, France

    Senior Python Consultant — Talan / SweetIT

    Within the SNCF Tele-diagnosis team, design and development of theMEQ Framework, a Python framework allowing business teams (SET2) to create, test, and deploy diagnostic pluginsfor rolling stock onboard equipment.

    Main achievements:
    - Design of a framework inhexagonal architecture(Python 3.14,Pydantic, Polars) exposing a declarative DSLfor filtering and association rules, usable without infrastructure expertise.
    - Development of adeveloper CLI(plugin scaffolding, local execution and testing, configuration validation), reducing onboarding time for business teams.
    - Development of theplugin-runner, an execution engine for plugins in AWS LambdawithSQSintegration, for scalable and resilient processing of tele-diagnosis streams.
    - Optimization ofdata pipelineswithPolars(LazyFrames, projection pushdown, lazy evaluation) on large volumes of equipment events.
    - Industrialization of distribution: publication onJFrog Artifactory, Windows installation scripts, quality tooling(uv, ruff, ty, pre-commit).
    - **Observability with Datadog**: Lambda and SQS queue dashboards, log-based metrics, pipeline alerting.
    -Technical documentation on Confluence(installation guides, tutorials, DSL reference) and support for user teams.
    - Agile methodology via **Jira**: user stories, sprint tracking, stakeholder reviews.

    Technical environment:Python 3.14, uv, Polars, Pydantic, AWS (Lambda, SQS, S3), PostgreSQL, Datadog, JFrog Artifactory, Jenkins, Jira, Confluence, Git, Gitlab
    Polars Python Amazon Web Services Pydantic Datadog
  • Visitdata
    Client Platform Maintainer
    SOFTWARE PUBLISHING
    April 2024 - Today (2 years and 4 months)
    Lyon, France

    Django / Angular Python Developer


    Maintainer of Visitdata's web platform (**Django REST Framework** /Angular**), enabling **data visualizationcollected and processed by theRCUinPowerBIformat, as well asuser account administration(rights management, sharing, source download).

    Main achievements:
    -Evolutionary and corrective maintenanceof theDjango REST Frameworkbackend:REST APIsexposing datasets processed by the RCU, management ofpermissionsand access.
    - Evolution of theAngularfrontend: interfaces for **account configuration**, **report sharing**, and **data source download**.
    - Integration and display of data inPowerBIformat (**report embedding**, management of workspaces and **access rights**).
    - Administration of **user management**: **fine-grained rights**, sharing between accounts, **access traceability**.

    Technical environment:Python, Django, Django REST Framework, Angular, PowerBI, PostgreSQL, Git
    Angular Django Django Rest Framework AWS S3 Python/Django
  • Talkto LLC
    Backend Python Django Developer
    SOFTWARE PUBLISHING
    June 2025 - November 2025 (5 months)
    Lyon, France

    Django Backend Takeover | Django Rest Framework for a Swift App, with Integrated AI Agent

    Subject to an NDA: until the application is on the market, I cannot detail its features.

    Key Achievement: Design of an AI agent integrated into the application

    -Conversational agentdeveloped withPydanticAI**, based on **GPT-4o(viaOpenRouter**), equipped with business **toolsto interact with the backend.
    - **RAG with partitioned Qdrant**: vector search with data isolation by partition.
    - Creation of anMCP(Model Context Protocol) to expose backend capabilities to the agent.

    Backend Stack:
    -Django DRF(Django Rest Framework)
    -Django Templatesfor the admin panel and testing tools
    - **PostGIS**: PostgreSQL with GEOS (C/C++ geometric calculation library)
    - **Redis**: cache and pub/sub
    - **Rust Axum**: WebSocket service supporting a large number of simultaneous connections, horizontally scalable via Redis pub/sub
    - **Django Celery**: asynchronous tasks
    Django Celery Rust PydanticAI Qdrant

Reviews

5.0

Out of 1 rating

ArnaudA

Arnaud

Jengo

Reviewed on 2/8/2025

Guillaume provided high-quality service on this project: technically very competent, adaptable and proactive, he demonstrated expert mastery of the Python / Django environment, while having good knowledge of other ecosystems (JS Server) and effectively met the project's needs and adapted to constraints. His enthusiastic and willing character made the collaboration smooth and pleasant. I highly recommend him!

Recommendations

Océan DrozOD
Rosendorn GavrielRG
Baptiste M.BM
Océan Droz and 2 other people have recommended Guillaume

These freelancer profiles also match your criteria

AgathaA

Agatha Frydrych

Backend Java Software Engineer

4.7

(3)

2

BaptisteB

Baptiste Duhen

Fullstack developer

4.6

(4)

5

AmedA

Amed Hamou

Senior Lead Developer

4

(2)

7

AudreyA

Audrey Champion

Web developer

4.3

(3)

4

Education

  • Software Architect
    42 Lyon
    2021
    École 42 Lyon — Formation développeur (peer-to-peer learning) L'école 42 est une école d'informatique novatrice basée sur l'apprentissage en peer-to-peer et la pédagogie par projets. Le cursus couvre aussi bien la programmation bas niveau (langage C, ASM) que le développement web full stack Python Django. Projets réalisés : - Application web full stack : backend Python Django / Django REST Framework (API REST, authentification, websockets) et frontend Angular / TypeScript, intégrant un jeu de pong multijoueur en temps réel entre amis. - Moteur graphique temps réel en C : rendu, gestion d'événements et optimisation des performances. - Moteur de raytracing en C : calculs d'intersections, gestion de la lumière et des matériaux. - Shell Unix en C : réimplémentation du fonctionnement de bash (parsing, exécution, pipes, redirections). - Programmation réseau en C++ : sockets et communication client/serveur à distance. Cette formation m'a apporté une compréhension profonde du fonctionnement des systèmes (mémoire, processus, réseau), un atout majeur pour concevoir des applications Django performantes et optimisées.

Skill set

Categories