AI Trainer @ DataAnnotation Tech
Mar 2024 — Present- Evaluate and improve state-of-the-art LLMs through code review, prompt testing, and software engineering analysis across C++, C, Rust, Python, and JavaScript.
- Create complex system instructions, coding prompts, and evaluation tasks to test model reasoning, code generation, debugging, and adherence to engineering best practices.
- Build isolated Docker-based testing environments, write validation scripts and golden solutions, and document evaluations with Markdown and LaTeX.