What I build
My interests revolve around natural interactions with digital services and AI as a tool to make it happen.
Motivations
My goal?
Create ways everyone can trust when interacting with technology. Be it for communicating, learning, or just playing around.
But here’s the thing.
How we connect with the digital world keeps evolving. We started with punch cards, then screens and keyboards. Then came the mouse and GUI. We got touchscreens, then pocket-sized devices. Now, it’s voice interactions.
More input. More ways to connect.
And with more input comes a deeper understanding of people.
But the big question: as things change, how important is data ownership in our future with digital assistants?
I want to build a world with the kind of tech I’d want my kids to use and live with for their entire lives.
Projects
Rico LeBot open-source (MIT) - GitHub
Real-time Voice Interface MVP
In November 2024, I've had the opportunity to participate in the OpenAI Builders Lab in Paris, where I was able to explore the potential of the Realtime API. I was amazed by how quickly I could build a prototype for a real-time web-based voice interface that uses function calls.
The initial prototype was focused on a cooking guide where you could ask for a recipe or ask for instructions on how to prepare a certain dish. It was an amazing and inspiring experience to me because I've spent the last 10 years building a similar technology. And it just works.
To go further, as I explored the possibilities of this prototype, it became clear there were some challenges to overcome before it could become a deployable product. That's why I created an open-source toolkit to address those problems.
Challenges and Solutions
-
WebSockets not well suited for longform connections: The official OpenAI Realtime API toolkit relies on WebSockets, which proved unstable for long-term sessions over HTTP. The OpenAI team suggested using WebRTC bridges for better stability. The toolkit implements WebRTC with a LiveKit integration.
-
Dynamic UI: I wanted a dynamic UI that could respond to user input in real-time. This meant connecting the function calls from the model to the front-end functions using remote procedure calls (RPC) over WebRTC. This brings the voice interface to life, allowing users to interact with different functions of the app seamlessly.
-
Architecture: A clear separation was needed between the web app's backend, the AI agent's backend, and the front-end. The toolkit achieves a lightweight design that is modular, has few dependencies, and is easy to use.
-
Roles: To quickly iterate on the user experience, you need to refine the prompts / instructions of the agent very frequently. That's implemented through an architecture where the 'roles' are separated from the code, and allows to add and modify them very quickly
Learn more in the blog post
Mojodex open-source (MIT) – GitHub
I've started to design and work on Mojodex as part of Hoomano's product with the team. It's been created hand-in-hand with our users. Initialy released as a SaaS product, it's been open-sourced because our clients wanted a way to build their own tool and create their future AI assets. That's what Mojodex provide them.
Mojodex has been designed as a fullstack AI assistant platform, bigger than Chat-based LLM projects.
If I can be of any help in your strategy of integrating AI in your business, don't hesitate to reach out
Instruct open-source (MIT) – GitHub
As a developer, I've felt like a lack of tools to interact with LLMs in an appropriate way. They are probabilistic interpretation engines using long sequences of texts as inputs.
Based on my own experience of writing LLMs instruction inside a project codebase, I've identified a few challenges:
- I wanted a way to create one input file per task which would adapt to many different LLMs.
- Templating is a must have
- Integrated in VS Code
Instruct is an open-source library and a VS Code extension to help developers integrate LLM-based features in their source code.
Sharing
Year | Event |
---|---|
2024 | Unapei – Le prendre soin au défi de l'IA |
2022 | Ecole LDLC > intervention de Xavier Basset Hoomano |
2020 | TEDx Que révèlent nos interactions avec les machines ? |
2019 | SIDO #SIDO2019: Xavier Basset présente HOOMANO |
2018 | Interview Xavier Basset |
2017 | AI StartUp Contest Xavier Basset (Hoomano), lauréat des Trophées des start-ups |
2016 | IBM Business Connect Caroussel du Louvre, 1000+ attendees |