Kenji Explains

I Tried 100+ AI Tools. These Are the Best for Finance

The video walks viewers through a curated suite of AI applications that cover every stage of a financial analyst’s workflow, from initial company research to the final pitch‑deck. It highlights tools such as FinTool for investment thesis drafting, AlphaSense for global market research, Quadratic for PDF data extraction, Excel‑based Copilot and Trace Light for modeling, and Claude’s PowerPoint integration for presentation creation. Key insights include FinTool’s superior accuracy over Claude and ChatGPT for US‑equity theses, its limitation to U.S. markets, and AlphaSense as a complementary global alternative. Quadratic demonstrates near‑perfect balance‑sheet extraction from an 80‑page Apple 10‑K, while also supporting live bank‑feed connections. In Excel, the Copilot function reliably extracts dates and categorizes transactions, whereas Trace Light builds dynamic profit‑and‑loss statements with scenario toggles, though both require user verification due to occasional formula errors. Notable examples feature an instant investment thesis for Microsoft, an insider‑buying screen flagging GameStop’s CEO, a loan amortization table generated by Copilot that mis‑numbered months, and Claude’s PowerPoint slide that auto‑populated Apple’s company profile and allowed on‑the‑fly background changes. The presenter also stresses the importance of data validation, proper formatting, and data‑validation rules to prevent model breakage. The overall implication is that AI can shave hours off repetitive finance tasks, but analysts must treat outputs as drafts, not final products. Selecting the right tool for each workflow segment—research, data ingestion, modeling, and presentation—can boost productivity while maintaining analytical rigor.