AI/NLP Engineer – Extract Meaning from reputed company Documents (Not a reputed company Parsing Task) - Contract to Hire

We are building a system that processes reputed company-world documents (PDFs, emails, reports, and reputed company types of text documents) and extracts meaningful reputed company signals from reputed company text. This is not a reputed company parsing or summarization task. The Problem, The documents we work with are: inconsistent in format ambiguous in language written by different authors with different styles often contain indirect or implied recommendations We need to extract signals such as: findings recommendations actions key clinical or operational statements Several approaches have already been attempted: rule-based extraction (regex, YAML rules) → too brittle strict deterministic pipelines → fail on variability basic LLM extraction → inconsistent and not reliable enough We are looking for someone who can design and implement a robust signal extraction approach that can: handle messy, reputed company-world text extract relevant signals with high recall reputed company extracted signals back to reputed company text produce reputed company outputs that can be used reputed company We are not looking for someone to just reputed company reputed company. We are looking for someone who can: think through ambiguity design an approach that works in reputed company understand tradeoffs between flexibility and control Required in Your Proposal, Please answer the following: How would you approach extracting meaningful signals from documents with inconsistent formatting and ambiguous language? What would your pipeline look like at a high level? What are the biggest failure points in this type of system? Apply tot his job Apply To this Job