About

Description: I’m looking for an experienced AI / LLM engineer who can build a working, reusable model or pipeline that takes a large body of text (~9,500 words) and automatically extracts the most important, meaningful sections — without being given any context or topic information. I’ve already tried using Python with model="gpt-4o-mini" and model="claude-sonnet-4-20250514", but both failed to consistently identify and extract key sections with high accuracy. I’m not looking for theoretical or research advice — I need an actual working solution that I can reuse for other similar datasets.