Blog
Notes from the work, and around it: privacy, red-teaming, quantization, and the odd weekend tool.
2026
- Oct 2026 Wrong, Lying or Scheming? Hallucination, Deception and Scheming in LLMs
- Sept 2026 How Much Can You Vibe? Building an iOS App Without Knowing Swift
- Apr 2026 How Vulnerable Are Multimodal AI Models to Simple Jailbreak Attacks?
2025
- Nov 2025 Where AI Agent Safety Benchmarks Stand Today
- May 2025 From SGD to DP-SGD: Reproducing the Foundations of Private Deep Learning Inception of Differential Privacy
- Mar 2025 The Art of Controlled Noise: Laplace and Exponential Mechanisms in Differential Privacy Inception of Differential Privacy
- Mar 2025 DP Guarantee in Action Inception of Differential Privacy
- Feb 2025 Differential Privacy!! But Why? Inception of Differential Privacy
2024
- Aug 2024 Exploring Llama.cpp with Llama Models
- Apr 2024 InterrogateLLM: In Search of Truth