Maxence P.

- Led end-to-end development of an LLM-powered security audit agent, built training datasets and designed evaluation methodology. - Designed agent workflows for threat modeling, vulnerability search, similarity search over past exploits, and static analysis integration, reducing manual audit effort. - Developed a bug bounty recommendation engine matching 500+ security researchers to optimal codebases, improving audit coverage and researcher engagement. - Created a code comprehension tool adopted by security researchers to navigate unfamiliar codebases, reducing onboarding time for new audits.

Cybersecurity skill acquisition to gain domain knowledge - Completed a security course - Published interviews with experts - Participated in competitions

- Led R&D of a real-time polyphonic guitar transcription system using a multi-head CNN, serving thousands of users in production. - Collected and curated 120h of audio training datasets; designed annotation workflows and data quality checks. - Optimized inference and signal processing pipelines to ~50 FPS on mobile (iOS & Android) in production.

- Built a PHP vulnerability scanner using deep learning to identify malicious behaviors and insecure patterns. - Designed deep neural network architectures (RNNs, LSTMs, CNNs) for behavior classification; evaluated models against labeled exploit datasets. - Assembled and labeled exploit datasets of 2 million samples; built data pipelines and evaluation harnesses for iterative model improvement.

Machine learning skill acquisition: - Published interviews of experts in the field (medical, computer vision, NLP, freelancers) - Followed high-quality courses (Andrew Ng, CS231n, CS224n, Hands-On Machine Learning book) - Built a portfolio with Kaggle competitions and summaries of research papers