Unclaimed profileBuilt from public GitHub and Hugging Face activity. Théophile has not joined Lounge yet.Claim it
Not on Lounge yet
Théophile Lafargue
founderFounder & CEO @ Sub Specie Aeternitatis: sovereign LLM inference over SMS, LoRa & satellite next. Apple Silicon inference optimization.
Public projects
mlx-dflash
3.34× faster inference on Apple Silicon: native MLX port of DFlash speculative decoding
Python
speech-to-speech-pipeline
A real-time, interruptible (barge-in) conversational AI pipeline (STT-LLM-TTS) running locally. Optimized for Apple Silicon (MLX).
Pythonapple-siliconasynciofaster-whisper
enigma-shell
An experimental web shell to control a full Linux OS (v86) with natural language via local LLMs.
JavaScriptagent-ai-cliailinux
opf-mlx
MLX implementation of OpenAI Privacy Filter with constrained BIOES Viterbi decoding and long-context banded attention
Pythonapple-siliconmlxner
instagram-unfollow-bot
Chrome extension to auto-unfollow Instagram accounts that don't follow you back, powered by your local LLM
grounding-engine
Keep a local LLM honest: it proposes validated typed deltas instead of mutating state. Anti-hallucination by construction. Zero deps, OpenAI-compatible.
JavaScript
Likely stack
OpenAILlamaOllamaWhisperDiffusionReact
