Data Extraction & LLM Analysis from SEC Filings

We are looking for an experienced Data Scientist / Machine Learning Engineer to help us build an automated pipeline that extracts, processes, and analyzes financial data from SEC.gov (EDGAR filings). The extracted data will then be transformed into structured reports and analyzed further using Large Language Models (LLMs) to generate actionable insights for our clients. Responsibilities Develop scripts to fetch and parse data from SEC EDGAR filings (10-D, ABS-EE, 424B5, servicer reports, etc.) Extract key structured metrics (e.g., Initial Pool Balance, Overcollateralization, Delinquencies, Charge-offs, Reserve Accounts, Weighted Avg FICO, etc.) Store, clean, and transform the extracted data into analysis-ready formats (CSV/SQL/Parquet). Build automated pipelines that generate monthly/quarterly trust performance reports. Integrate with LLMs to summarize results into plain-English business reports. Ensure reproducibility, scalability, and accuracy of the data pipeline. Requirements Strong Python skills (Pandas, NumPy, BeautifulSoup/lxml, Regex) Familiarity with SEC.gov EDGAR filings (ABS, 10-D, 424B5/424H prospectuses, etc.) Experience with data engineering and automation (Airflow/Prefect/Dagster preferred but not required) Knowledge of LLMs (OpenAI, Anthropic, Cohere, etc.) and how to use them for summarization/insights Strong understanding of finance, ABS securitization, or structured credit analysis is a plus Deliverables Automated pipeline to fetch and parse SEC filings Extracted datasets with relevant financial metrics Report generation module (CSV/Excel/PDF) LLM-based analysis report (e.g., “Cumulative Performance Summary,” trends, erosion of OC cushion, delinquency growth, etc.) Documentation for reproducibility Why Work With Us? We are building cutting-edge financial analytics solutions using a mix of traditional data engineering + modern LLMs. You will work directly with a team experienced in ML, financial modeling, and reporting automation.