Automate a business workflow with AI
Map a repetitive process, select the right model and connect a secure agent or RAG system to the tools your team already uses.
Freelance Software Engineer — Full-stack, Integrations & Applied AI
I help startups, SMEs and technical teams design, strengthen or modernize software products—from a backend or SaaS feature to a mobile application, an IoT platform, an AI workflow or technical C++/Python software. I am available for fully remote work worldwide.
My public work includes software modernization at the Tunnel Study Center (CETU), AI and systems experiments, and open repositories that make parts of the approach checkable. Confidential work remains protected.
I bring a product and delivery perspective: I care about the business problem, the delivery risk and the result, not only the stack.
Choose a focused intervention or combine them into a complete delivery.
Map a repetitive process, select the right model and connect a secure agent or RAG system to the tools your team already uses.
Design APIs, microservices and integrations that stay understandable and reliable as the product and team grow.
Improve calculation, simulation, rendering or migration software with a pragmatic mix of C++, Qt, Python, performance work and UX improvements.
Build a customer-facing application or internal tool end to end, from data model and APIs to responsive web or mobile interfaces.
Turn a technical idea into a demonstrable prototype using computer vision, machine learning, robotics, IoT data flows or simulation.
Supported lighting sizing for French tunnels through safety-critical software using C++, Qt, Python, and embedded Linux.
Public reference + pictures available.
Enabled multiplayer gameplay across 2D and 3D worlds through a game engine with Unreal-like capabilities using C++.
Work in progress.
Automated workflows across web and mobile interfaces through an n8n-like application using FastAPI, Python, Next.js, Tailwind, React Native, and React.
Enabled real-time competition between autonomous teams through a distributed multiplayer strategy game with real-time visualization using C++ and Python.
Rendered complex 3D scenes with physically based lighting through a high-performance distributed ray tracer using C++.
Accelerated experimentation across language, vision, and reinforcement-learning models through reusable training and evaluation pipelines using Python and PyTorch.
Expanded access to neuromorphic computing through an open-source course with hands-on spiking neural network exercises using Python and PyTorch.
Deepened practical understanding of LLM architecture through a GPT-2 implementation and training pipeline from scratch using Python and PyTorch.
Participated in the SWERC 2025 regional programming contest through team-based algorithmic problem solving using C++ and Python.
Selected public projects are shown above. Confidential work can be discussed without sharing protected code, data, or screenshots.
Read the case studiesHave a project in mind?
Send me the context, the current bottleneck and your target date. I’ll reply with a concrete first proposal.