Datafied Investment Reasoning in an Emerging VC Firm: Designing and Studying AI along the Investment Process (Cooperation with Drumbeat Capital)

Thesis (BA/MA/GR)

Advisor: Maximilian Harl (maximilian.harl@tum.de) in cooperation with Drumbeat Capital

CONTEXT

CONTEXT The investment thesis, a forward-looking and revisable proposition about where and why venture scale returns can be earned, is the central reasoning artefact of a venture capital (VC) firm, yet in practice it stays largely tacit and is rarely made explicit. A new generation of “AI-native” managers is building firms around data-driven reasoning, deploying AI across the investment process and in upstream technology mapping and evaluation. How this reasoning is actually done when a firm is built from scratch around data and AI, and whether datafying it captures or changes the investor’s tacit judgement, is barely understood. In cooperation with Drumbeat Capital, an emerging VC firm being built from the ground up, the student gains rare real-time access to this process. The task is to design and prototype an AI/data driven artefact for one part of it and study how the tool reshapes the firm’s reasoning: what it makes explicit, and what tacit “know-how” resists codification. Grounded in IS theory on datafication and human/AI augmentation, the thesis can take one of three directions, chosen together with the firm and the advisor:

Possible Directions

• AI along the deal funnel. Prototype how AI can support the full VC lifecycle and study which stages datafication genuinely improves.
• Investment thesis-generation engine. Build an engine that surfaces technological bottlenecks, tracks patents and emerging technologies, and proposes investable theses, and study how datafying thesis genesis captures or changes tacit know-how.
• Founder and team evaluation. Treat the first call as the decisive moment for judging founders, design AI-supported instruments for team due diligence, and study how far this judgement can be datafied.

Task

• Review the literature on VC reasoning and on the datafication of expert decision-making  
• Create a reference architecture for Drumbeat’s investment process and tool stack
• Design and prototype a AI/data-driven artefact using state-of-the-art tools and available data sources
• Evaluate it in the live firm setting and analyse how it changes behavior (what it captures, what it misses)
• Derive design principles and a conceptual framework for datafied investment thesis resoning in emerging VC firms

Requirements

• High degree of autonomy and individual responsibility
• Strong communication skills for interacting with practitioners
• Interest in current IS topics, venture capital, private equity, and strategic decision-making
• Experience in & willingness to conduct scientific studies, analyze qualitative and quantitative data, and learn about scientific writing
• Structured, reliable, and self-motivated work style

FURTHER INFORMATION

This thesis is offered in cooperation with Drumbeat Capital; close collaboration with the firm is expected. The scope can be adapted to your interests, both the direction (1, 2, or 3) and the depth (very good BA-student or MA). The thesis can be written in English or German. If you have any further questions, do not hesitate to contact me directly. Please send your application, including our application form, current transcript of records from TUMonline, and CV to maximilian.harl@tum.de. Please note that we can only consider applications with complete documents.