About Me

I'm interested in ML, AI, Security, and Privacy. I'm currently a Senior Machine Learning Engineer at Meta, where for the last few years I've been working on incubating the use of ML and LLM deployments in the Privacy and Security orgs.

My latest work has been focused on mitigating security and reliability risks in GenAI applications. A lot of this work has been released open-source into Meta's Llama ecosystem. Projects I work on include CyberSecEval, PromptGuard, LlamaGuard 3, and the Llama Stack.

Before getting involved with GenAI and LLMs, I worked on traditional ML, particularly large-scale recommendation systems. I built the first version of Instagram's feed recommendations ranking model and also worked on Facebook's search engine.


Updates

Feb 2025
Speaking Engagement
Will be attending and hosting a workshop and demo session at AI Security Forum in Paris.
Aug 2024
Speaking Engagement
Spoke at DefCon AI Village — "Evaluations and Guardrails against Prompt Injection attacks on LLM-powered applications."
Aug 2024
Speaking Engagement
Hosted a workshop at the AI Security Forum .
Jul 2024
New Paper
Llama 3 Herd of Models. Our team worked on cybersecurity risk measurements of the models and the system-level safety models.
Jul 2024
Product Launch
Launched PromptGuard and LlamaGuard 3, part of Meta's system-level safety stack for Llama.