Skip to content

rubenmst/TalentVest-MVP-

Repository files navigation

TalentVest – MVP: Empowering Education through ISAs

A FinTech MVP platform connecting students and investors via tradable Income Share Agreements (ISAs). Developed for the MSc Business Analytics & Management programm and its FinTech course at Rotterdam School of Management (RSM). The goal is to allow more students to access higher education while creating a new impact assest class for Investors.


Overview

TalentVest enables:

  • both parties (Investors and Students) to connect in a single platform
  • Students to apply for funding and update their academic journey in the student portal/dashboard
  • Investors to browse, fund, and trade ISA shares in the Investor Dashboard
  • Portfolio tracking with ML-based estimates used for ISA right valuation and profit projections to aim to increase transparency
  • increased liquidity for Investors by offering a Trading functionality, which results in higher investment incentives

All data used is simulated. The ISA valuation engine leverages simple regression models to project salary, growth, and discounting. At a later scaling stage, a real-life data will be collected and utilized to create a unqiue student database as a foundation for the ML models and ISA valuation.


Features

Student Portal

  • Application form + document upload to request funding and be listed on the website + a Questionarry with further a further personal fit assessment is in planning
  • Academic progress and quarterly update form is required to continously update a students profil, as students need to be transparent about their academic journey
  • Privacy and ethic standards are always considered by TalentVest

Investor Dashboard

  • Explore student funding opportunities
  • 1st stage: High-school graduates that applied for a University but require the money
  • Student Details: Several details and variables are displayed about the student and upcoming studys
  • Financials: yieling finanical variables, such as required funding, ML estimates for salary, discount rate,etc.
  • Funding a flexible amount directly in the platform
  • Navigate and filtered across student to fund and invest in

ISA Trading Module

  • Browse existing funded students available for resale: submit offers on secondary market ISAs
  • 2nd stage: The ISA rights for Unversity Students or already employed students can be listed and traded by Investors
  • Student Details: similiar as before and updated quarterly
  • Finanicals: similar as before and updated quarterly, updated ML estimates and ISA valuation as a guideline for Investors
  • Offer: Investors can submit an offer (price) to the other Investors for the ISA rights of the student

Portfolio Overview

  • Listing of the ISA rights that are hold by an Investor
  • Detailed View and detailed Financials as before
  • Including NPV and ROI calculations per student to track process and profitability
  • Aggregated financial visualizations
  • Option to list students in the ISA Trading Module to other Investors

Tech Stack

  • Frontend: Streamlit (entirely Python-based)
  • Backend:
    • pandas, scikit-learn, matplotlib
    • ISA logic implemented via custom Python modules
  • Data: Fully simulated to show feasibility of salary, discount rate (risk), and salary growth prediction

Project Structure

TalentVest_MVP/
├── app.py                            # Streamlit UI & navigation
├── requirements.txt                  # bundeled libraries/packages used  
├── README.md                         # the file you are reading right now  
├── .gitignore
│
├── data/                             # students.csv, trading.csv, portfolio_sample.csv (simulated profiles with ML estimates)
├── static/                           # Logo and image assets 
├── models/                           # Trained salary/discount models
│
├── generate_students.py              # Student profile generator 
├── train_salary_models.py            # Model training
├── predict_targets_on_new_datasets.py   # predict salary, growth, and discount rate  
├── recalculate_isa_value_from_predictions.py # update ISA right valuation for simulated data sets used 
├── portfolio.py                      # Portfolio aggregation logic
├── npv.py                            # Discounting & NPV helper functions
├── dataset.py                        # ML feature schema and transformations

ISA Valuation Engine

Trains and applies ML models to estimate:

  • predicted_salary
  • salary_growth
  • discount_rate

These values feed into:

  • ISA Share (%), NPV, ROI, Estimated Profit

Thereby, increase transaprency and liquidity is provided by TalentVest.

All calculations done with:

  • Discounted cash flow logic
  • Assumed 15% ISA share fixed across investments across 10 years pay-pack period from Students to Investors
  • Simulated Data sets to show feasibility
  • Basic OLS and Lasso model including interaction effects were constructed, more sophisticated Neural Networks will be utilized later

Installation Guide

  1. Clone the Repository:
git clone https://github.com/rubenmst/TalentVest-MVP-.git
cd TalentVest-MVP-
  1. Install Requirements:
pip install -r requirements.txt
  1. Run the App:
streamlit run app.py

MVP Demo Video

Video: https://drive.google.com/file/d/1TgwYixqvp8guzAwtrgo5lWFDLrGSg-SO/view?usp=sharing Folder: (https://drive.google.com/drive/u/0/folders/1lLRTIlLR1c11MVrKoXgqKVFiXFrml0Db)


Academic Context

Built as part of:

  • Course: Business Models & Applications in FinTech (BM26BAM)
  • Programm Business Analytics and Management
  • University: RSM, Erasmus University
  • Assignment 2: MVP implementation of business idea - TalentVest

This MVP reflects the execution of Assignment 1 (Business Plan) into a working, testable prototype.


Disclaimer

This is an educational project. All data is simulated and predictions are hypothetical. This is not financial advice.


Future Improvements

  • Demo with a first cohort of students
  • Add database integration (PostgreSQL)
  • Track student data (anonyously) to build real-life data set for ISA right valuation in the long-term
  • Enrich databank by connecting ISA logic to real-world salary APIs / market data
  • Deploy on Streamlit Cloud or Heroku
  • Add authentication for separated Investor and Student Access
  • Improve ML models with hyperparameter tuning

License

MIT License – Free to use and modify for educational or demo purposes.


Maintainer: Ruben Maxim Stauch | MSc BAM | RSM

About

MVP platform to match students and investors via ISAs – FinTech Assignment @ RSM

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages