I am hiring a Python / LLM research engineer for an academic project that adapts an existing research codebase on LLM safety evaluation to online chatbots. The work includes understanding a nontrivial codebase, building a reproducible prompt-response collection pipeline, organizing outputs, and helping produce initial analyses for a paper. This is not a basic scripting task. I am looking for someone with strong Python skills, experience with LLM evaluation workflows, and comfort working with research-grade code and analysis. The goal is to complete the core implementation and first empirical outputs within one month. Full project materials will be shared with shortlisted candidates.