~/blog

Perspectives on AI, security, and systems.

SDR-Bench: How We Built the First Personalization Benchmark for LLM Agents

Coming soon — a deep-dive into the design decisions behind SDR-Bench, the first benchmark for evaluating personalization in LLM-based sales agents.

2025-08-01h4shk4t

ReEdit: Multimodal Exemplar-Based Image Editing with Diffusion Models

ReEdit is an efficient end-to-end optimization-free framework for exemplar-based image editing. Unlike existing approaches, it doesn't require fine-tuning or optimization during inference time.

2025-05-30h4shk4t