Matthew Kerr

Built SQLite and Python/SQL data pipelines to consolidate and reconcile holdings, transactions, financial statements, SEC filings, and market/options data for portfolio valuation and analysis. Analyzed portfolio performance and exposure using Python, SQL, and Excel, highlighting concentration risks to support allocation decisions. Developed RAG-enabled multi-agent research workflows using LLM APIs, REST APIs, and MCP, and delivered auditable insights with validation and Power BI reporting.

Developed an end-to-end equity research workflow covering SEC and market-data ingestion, three-statement modeling, DCF valuation, scenario analysis, and reconciliation into investment reports. Automated research and valuation steps by combining deterministic Python calculations with LLM interpretation. Added validation checks to flag incorrect or unsupported inputs before downstream use.

Built a portfolio-aware covered-call agent that ranks opportunities using upstream valuation outputs and live stock and options-chain data. Implemented probability modeling (including Black-Scholes-based and analyst-adjusted probabilities) and dividend early-assignment risk to compare calls by expected return versus assignment risk. Integrated market data via Alpaca and MCP and scored calls deterministically under portfolio constraints.

Created a SQLite-backed ETL pipeline to consolidate account holdings, transactions, daily prices, and benchmark holdings. Reconstructed historical holdings with time-versioned ETF and sector mappings to compute time-weighted returns and sector attribution. Produced consolidated Excel outputs comparing SPY/sector ETFs and calculating beta and dollar-based rebalancing targets.