Need Co-Founders--Industrializing the "Investment Committee" for the $1.7T Private Credit Market
I am a veteran credit fund professional (20+ years, high-level distressed/special situations) building "Howard." Howard is an AI agent that doesn't just read data; it underwrites risk. We are automating the cognitive workflow of a credit analyst—from scraping BDC filings to modeling downside scenarios.
Need A Technical Lead / Data Scientist who excels at the "dirty work" of data engineering—ingestion, cleaning and pipelining. I need someone who can build the initial data infrastructure (Python/SQL/Vector DBs) that feeds our LLM reasoning engine.
Async timing systems usually break or succeed based on how time boundaries are defined, not the tooling itself.
A useful way to test them is to look at:
Curious — are you optimizing more for reducing decision latency, or for reducing rework caused by decisions made too late?
I did something quite similar for 8-K report analysis, though likely on a lesser scale. Pipeline complexity can significantly vary depends on the shares of structured/unstructured and numerical/verbal data that need to be prepared, and response time constraints. Also, do you own your LLM or fine-tune the external model?
This resonates! I design virtual workspaces that solve exactly this. Doing free setups this week if you want to test with your team