Hey Palantir Community!
Munich-based builder here — first post, so bear with me! 
The Project
I’m building a code repository and data pipeline to track AI transformation across the top 40 reinsurance companies worldwide — pulling signals straight out of financial and annual reports and turning them into a living dashboard of who’s talking about AI, who’s actually shipping it, and how that shifts year over year.
The Approach
Under the hood it’s an ontology-management problem as much as a data problem: structuring messy annual-report data into something queryable, then layering NLP extraction and benchmarking on top. Happy to turn this into a mini workshop if others are wrestling with similar report-parsing challenges.
Looking For
Fellow builders working on similar pipeline-builder style projects and code repositories, folks with ideas on report parsing, NLP extraction, or benchmarking approaches, and advisors who’ve wrangled messy unstructured financial data before.
Let’s Build Together
Happy to share more on the approach so far, or just swap notes on what’s worked (or spectacularly failed) for you.
Excited to be here — let’s make something great! 