Skip to content
View eauchs's full-sized avatar

Block or report eauchs

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
eauchs/README.md

Hi, I'm Théophile Lafargue 👋

I build and deploy high-performance AI systems, primarily focused on quantitative finance.

My work bridges the gap between deep technical implementation and high-level strategic vision. This led me to architect the "Alpha Narratif" market simulator at Amundi and to file my first patent for a stateful AI system.

I operate on two modes:

  • Professional R&D: Building end-to-end AI prototypes for quantitative research at Europe's leading asset manager.
  • Self-Directed Research: Running a dedicated R&D homelab where I master the full MLOps stack, from bare-metal hardware (Arch Linux, MLX) to real-time, interruptible conversational pipelines.

Featured R&D Projects

Here are the public projects that demonstrate the types of systems I build.

Project Category Key Concept
patent-low-bandwidth-ai Patent / Backend (Patent Pending) A hybrid RAG backend that enables stateful AI conversations over low-bandwidth networks like SMS.
gui-agent AI Agents A two-layer autonomous agent for macOS, separating visual perception (VLM) from strategic decision-making (LLM).
speech-to-speech-pipeline MLOps / Real-Time A low-latency (STT-LLM-TTS) conversational pipeline with barge-in, optimized for Apple Silicon (MLX).
enigma-shell Full-Stack AI An experimental web shell to control a full Linux VM (v86) using natural language via local LLMs (Ollama).

My Technical Focus

  • AI for Quantitative Finance: Modeling market reflexivity, information warfare, and narrative impact.
  • AI Agent Architecture: Designing multi-layer systems (perception, strategy, execution).
  • MLOps & Systems: Building high-performance, real-time data pipelines (asyncio, queues).
  • Full-Stack Prototyping: Python (Flask, PyAutoGUI), React (TypeScript), MLX.

Connect with Me

You can find the full story of my professional experience on my LinkedIn profile.

My LinkedIn Profile My Hugging Face Profile

Pinned Loading

  1. enigma-shell enigma-shell Public

    An experimental web shell to control a full Linux OS (v86) with natural language via local LLMs.

    JavaScript 3

  2. speech-to-speech-pipeline speech-to-speech-pipeline Public

    A real-time, interruptible (barge-in) conversational AI pipeline (STT-LLM-TTS) running locally. Optimized for Apple Silicon (MLX).

    Python 2

  3. gui-agent gui-agent Public

    A two-layer GUI agent for macOS. A VLM (Vision Language Model) handles perception, while a separate LLM (Qwen) manages high-level strategy and decision-making. Built with Python, OpenAI API, and Py…

    Python

  4. patent-low-bandwidth-ai patent-low-bandwidth-ai Public

    Backend for my 'Stateful AI over Low-Bandwidth Networks' patent (FR2511116). A hybrid RAG pipeline (SmolDocling, ChromaDB, Reranker) with SMS support, separating local VLM (perception) from the mai…

    Python