AI Model Alignment Specialist (Post-Training)
I performed instruction tuning and evaluation benchmark setup on large language models using Hugging Face. My responsibilities included aligning AI model outputs with precise legal and linguistic requirements to bridge machine learning with legal expertise. I contributed to ongoing AI model alignment research and prompt engineering. • Assisted in model output evaluation for safety and logic.• Focused on legal reasoning encoded via model alignment.• Tuned LLMs based on expert-crafted instructions.• Employed RLHF and evaluation tools via Hugging Face.