Skip to content
View schuelting15's full-sized avatar

Block or report schuelting15

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
schuelting15/README.md

Dennis Schülting

I'm doing an MSc double degree at NHH (Bergen) and the University of Mannheim, and I interned at nFront Ventures in summer 2026. I care about venture capital and the data work behind it: sourcing, screening, and testing what actually predicts outcomes. I build the tools for that myself, from asset-pricing pipelines in Python and R to automated deal-sourcing workflows.

Focus: venture capital · data analytics · automation

Tech stack: Python (pandas, NumPy, scikit-learn, matplotlib) · R · TypeScript/React · n8n · Claude API · Git

Selected projects

  • The Disruption Score: links 10-K business descriptions to venture-backed start-ups by text similarity. It tests, with a pre-registered design that avoids look-ahead bias, whether VC funding into a firm's product space predicts its stock returns (Python, R).
  • Pulse: a concept prototype of a quarterly founder check-in for a VC platform team. It handles validated impact KPIs with an audit trail and routes founder requests to the best-matching helper (TypeScript/React, 108 unit tests).
  • A/B testing and multi-armed bandits: cleans and analyses about 60,000 app sessions from a three-arm recommendation experiment, then simulates ε-greedy, UCB1 and Thompson Sampling. Thompson Sampling cuts regret against a fixed split by about 97% (R, synthetic data).

LinkedIn

Pinned Loading

  1. ab-testing-bandits ab-testing-bandits Public

    A/B/C test and multi-armed bandits (epsilon-greedy, UCB1, Thompson Sampling) for a ride-hailing app's recommendation strategies. R, synthetic data.

    R

  2. pulse-founder-checkin pulse-founder-checkin Public

    Concept prototype: quarterly founder check-in for a VC platform team. Impact KPIs with validation and audit log, request-to-helper matching. React/TypeScript, invented data.

    TypeScript

  3. vc-disruption-score vc-disruption-score Public

    Does VC funding into a public firm's product space predict its stock returns? Text-similarity score (10-K vs. start-ups), pre-registered asset-pricing tests in Python & R.

    Python