Pierre Guillaume
About
Pierre Guillaume is from France. Pierre is currently Co-founder, CTO at OLENO, located in Paris et périphérie. Pierre also works as Co-founder at Vauban AI, a job Pierre has held since Jan 2024. Another title Pierre currently holds is Software Engineer at Criteo.
Pierre Guillaume's contact information is available for free on finalscout.com, a web-based professional networking database with more than 500 million business contacts and 200 million company profiles.
Pierre Guillaume's current jobs
+ Built a sovereign AI-as-a-Service platform (Vauban Cloud) from the ground up on proprietary GPU infrastructure, ensuring full data privacy and compliance + Deployed and served production-grade AI workloads directly on self-hosted infrastructure, ensuring data sovereignty without reliance on public cloud providers + Developed Himeji, a local RAG-based AI assistant, enabling secure "Chat with your files" capabilities for enterprise and personal use + Engineered high-performance inference pipelines for open-weights models (Qwen, Mistral, etc.) using state-of-the-art engines (vLLM, TGI, TensorRT-LLM, TextSynth ...) to maximize throughput and minimize latency + Full Stack Development (Python / JS / Quasar) + End-to-end Product Engineering: Translated technical capabilities into user-centric features, overseeing the product lifecycle from conception to release
+ Working on Machine Learning bidding models + Working with huge dataset and high performance constraints for inference + Working with Google chrome on Privacy sandbox
+ Working on Audio classification + Set up Speaker identification 2022 recruitment subject (Kaggle competition / Speaker classification / Speaker verification) + Research on Hate speech and toxic comment detection (NLP) + LSE winter days talk: « Welcome to the world of Transformers: from moderation to code generation » (Gconf youtube channel: https://www.youtube.com/watch?v=KeX6q_s3Z-s&t=4992s) + Publication in the workshop DL4NLP at EGC 2022 (« Hate speech and toxic comment detection using Transformers ») + Publication at EGC 2023 and RNTI: https://editions-rnti.fr/?inprocid=1002807 ("A benchmark for toxic comment classification on Civil Comments dataset")
+ Machine Learning engineer at Alsid (acquired by Tenable) + Implementing an end-to-end ML solution for Anomaly Detection in large-scale Active Directory systems (Python / Go / C++) + Data Mining: Extracted user data by recursively crawling through Microsoft Active Directory architecture + Big Data Feature Engineering (30 million features) and Unsupervised Learning (Clustering, Autoencoders, PCA, Anomaly detection)
Internship at Thiga. SEO Wordpress. HTML/CSS/Javascript.
Internship at NeuroSpin in CEA. Development in python/bash. Servers setup. Lightweight Directory Access Protocol (LDAP) Image/data processing.