Freelancer profile translated to English.

Description

With over 13 years of experience in mobile development, I am currently a Lead Android Developer specializing in the design of high-value native Android applications. My background has allowed me to lead large-scale projects for renowned companies like Geodis, combining performance, software quality, and advanced user experience.

Expert in Kotlin, Java, and Jetpack Compose, I integrate best practices in architecture (MVVM, Clean Architecture, Coroutines, RxJava) and agile methodologies to ensure reliable and scalable solutions.

In parallel, I am developing expertise in Generative Artificial Intelligence applied to mobile: integration of LLMs (GPT, RAG, NLP), image processing with Computer Vision & OpenCV, and deployment of AI models optimized for mobile (TensorFlow Lite, PyTorch Mobile).

My goal: to support companies in creating intelligent native Android applications capable of leveraging the latest AI advancements to transform user experience and generate business value.

Industry field of expertise

Languages

French
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Paris (up to 50km)

LOCASMART
Lead Mobile Developer
SOFTWARE PUBLISHING
February 2025 - October 2025 (8 months)
Paris, France
Project: Ecosystem for the digitalization of car rental agencies
Development and maintenance of a complete Flutter ecosystem for the digitalization of car rental agencies, consisting of two interconnected applications:
- LS-Agency — Management application for agencies
Flutter application (mobile and web) allowing agencies to manage their rental business:
Management of customers, vehicles, reservations, payments, invoices, and expenses.
Dynamic dashboard for fleet and performance monitoring.
Integration of REST APIs for real-time data synchronization.
Use of Provider for state management and code structuring.
Advanced features: electronic signature, document scanning, PDF contract generation.
Secure authentication, user roles (admin, agent), and push notifications.
- LS-Renty — Customer application for car rental
Flutter application for end customers, customizable according to agency preferences:
Consultation of the vehicle catalog, filtering by category, price, or availability.
Instant booking, order tracking, and online payment.
Automated vehicle access via digital key and QR Code.
User profile management, reservation history, and notifications.
Modern, fluid, and responsive interface
Flutter Dart REST APIs Firebase Cloud Messaging Android Studio
LOCASMART
Lead Mobile / AI Developer
SOFTWARE PUBLISHING
January 2024 - February 2025 (1 year and 1 month)
Paris, France
Design and deployment of a multi-channel intelligent chatbot solution (Web, WhatsApp, Telegram) based on open-source LLMs and RAG architectures to automate responses to internal and external users (Support, HR, Finance, etc.).
Project management of mobile / AI as Tech Lead, coordination of a team of 3 developers (frontend/backend) + 1 QA, using Agile methodology (Scrum).
Implementation of a RAG pipeline with vectorization (pgvector), semantic search (Elasticsearch), contextual re-injection, and response scoring.
Development of a Python/FastAPI backend exposing REST APIs and
Integration and orchestration of open-source LLMs (LLaMA, Mistral via Ollama) with advanced prompt management, dynamic reformulation, Zero/Few-shot strategy.
Automation of linguistic pre-processing: spell checking, language detection, lexical simplification, transformation of user queries.
Extraction of data from documents from heterogeneous sources (HTML, PDF, XLS, CSV) for enrichment of the documentary base, with integrated cleaning and automated data classification pipelines.
Support for dynamic multilingualism configurable by client with automatic fallback.
Design of advanced Prompt Engineering strategies, including dynamic templates, role injection, business contextualization, hallucination reduction, and generation guidance.
Flutter administration interface (BLoC) with multi-tenant management, analytics, configuration of models used, complete bot customization.
Modular pre/post-processing pipeline, chaining: vectorization → search → selection → LLM → re-ranking → post-processing.
Technical environment: Python, FastAPI, Docker, Flutter, PostgreSQL, pgvector, Elasticsearch, LangChain, Mistral, LLaMA, Ollama, PyPDF2, pdfplumber, pytesseract, Dio, GitLab CI/CD, Jira, Confluence, Hugging Face.
Android Generative AI Langchain Ollama Flutter
Geodis
Lead Mobile Developer
TRANSPORTATION
October 2021 - December 2023 (2 years and 2 months)
Paris, France
Project: MobiScan

Development and deployment of an innovative timekeeping application for Geodis agencies. It integrates a laser, whether built-in or via Bluetooth, and offers the ability to capture package photos for instant updates with the Customer Relations department.

Project: Mobicop

Maintenance and feature additions to an application for drivers making delivery rounds for GEODIS Distribution & Express. The application offers a variety of features:

Project: MobiPrep

Maintenance and feature additions to an application designed to facilitate package preparation for rounds by agents. It allows for optimizing package distribution based on truck type, while also considering low-emission zones.
Android Kotlin Jetpack Compose Java Mobile Development