🤖Webinar - From Theory to Practice: Real-World Applications of AI in Asset-Based Finance - Register here

We are looking for a highly motivated and creative Ph.D. candidate interested in long term hierarchical and grouped time series forecasting. This Ph.D. project aims to predict cash flows of a private/illiquid portfolio (e.g., Mortgages, SME loans, consumer loans) across multiple geographies and industry sectors. During the project, you will closely collaborate with industry and a doctoral training network spread throughout Europe, including extended research stays abroad. 

The successful applicant will join the Data Science Team within Cardo AI, with the degree conferred in collaboration with the University of Kaiserslautern-Landau (RPTU)


This Ph.D. position is one of 2 positions at Cardo AI in the context of the international Marie Skłodowska-Curie Actions project DIGITAL. For the general description of DIGITAL and the Ph.D. positions, please check the official project webpage.  

DIGITAL’s main goal?

To significantly advance the methodologies and business models for Digital Finance through the use of five interconnected research objectives:

  1. Ensure sufficient data quality to contribute to the EU’s efforts of building a single digital market for data 
  2. Address deployment issues of complex artificial intelligence models for real-world financial problems 
  3. Validate the utility of state-of-the-art explainable artificial intelligence (XAI) algorithms to financial applications and extend existing frameworks 
  4. Design risk management tools concerning the applications of the Blockchain technology in Finance 
  5. Simulate financial markets and evaluate products with a sustainability component 

The outcome of this individual research project will contribute to the expanding body of knowledge concerning the applications of cutting-edge machine learning and artificial intelligence techniques to traditional financial problems. Specifically, the first phase of the project will concentrate on missing value imputation for loan payment time series, while the second phase will adopt a more general predictive approach, that of grouped time series forecasting, possibly incorporating the first step.

The challenge

Data incompleteness and inconsistency in institutional loan portfolios significantly hampers the accuracy and effectiveness of financial models. Candidates will tackle issues arising from missing information within loan datasets, e.g., repayment and loan status, employing a variety of advanced missing value imputation techniques.

The primary research objective is to develop a sophisticated machine learning tool capable of grouped time series forecasting for private debt portfolios that span diverse geographies and sectors. By leveraging both public and proprietary data, the successful candidate will work on refining and advancing financial modeling techniques. This initiative aims to provide institutional investors with more accurate and useful insights to refine investment strategies and enhance model robustness, thereby making significant contributions to financial analytics and effective investment management.

Your profile

We look for a highly motivated, enthusiastic researcher who is driven by curiosity and has/is: 

General skills: 

Project-specific skills: 

Interested and motivated candidates are encouraged to apply, even when not yet possessing all desired skills. Through self-driven learning and doctoral training, you will be able to develop relevant skills on the job.  

Our offer

Benefits offered as part of this position include: 

  1. Living allowance of EUR 2171/person month (net), Mobility allowance of EUR 600/person month, Family allowance of EUR 660/person month. More details on the allowances can be found here.
  2. Company-provided laptop, monitor, keyboard and mouse, flexible working time, work from home or office, internal and external education as per the set OKR/KPIs guided by mentors, meal vouchers of EUR 5 for each working day.
  3. This PhD position includes two research stays at industrial partners. The first research stay will be carried out at Humboldt University of Berlin (Germany), under the supervision of Prof. Dr. Wolfgang Härdle. It is scheduled to start on Month 12 and last for 18 months, during which the candidate will be exposed to various bodies of research on Fintech innovations and their applications. The second research stay will be carried out at Athena Research Centre in Athens (Greece), under the supervision of Prof. Dr. Ioannis Emiris. It is scheduled to start on Month 33 and last for 4 months, during which the candidate will be exposed to applied industry-research in a world-leading research center and make use of its infrastructure.
    1. Any potential change from the initial plan regarding research stays will be dully notified to candidates and reflected in the job advert description.

How to apply

Are you interested to be part of our team? Please submit your application, and include:  

Please ensure that your application is submitted by the deadline and note that we will start conducting interviews with short-listed candidates starting from April 2024; however, the application deadline is the 5th of July 2024. Do not make submissions via email as they will not be considered.  

Additional information can be acquired via email from Dr. Gennaro Di Brino (gennaro.dibrino@cardoai.com) with a cc to Dr. Jorg R. Osterrieder (jorg.osterrieder@utwente.nl). 

Diversity and Inclusion

We encourage applications from minorities and underrepresented groups to enrich our diverse academic community. Candidates will be selected on the basis of their competence and ability, and all applicants will be given equal opportunities. We acknowledge the importance of diversity and inclusion for innovation and excellence in digital finance research.

About the department

The Cardo AI Data Science Team includes people with world-class backgrounds in hard science and engineering. The team is structured around our main areas of activity, and it consists of three units: Pricing and Optimization, Tabular Data Modeling, and Unstructured Data Modeling. We work on a diverse range of problems, from Portfolio Optimization (loans, notes, etc.) to Probabilistic Classification (e.g., probability of default), from Time Series Forecasting to Language Modelling. Our mission is to assist decision makers in structured finance markets with the latest, most reliable modeling techniques available in machine learning and quantitative finance research.

About the organization

Cardo AI is a Milan-based software, data, and intelligence company innovating the asset based finance and private credit market. Founded in 2018, our solutions have been designed to help investors, issuers, servicers and lenders unlock opportunity from complexity and transact faster, better, and with lower costs. As of January 2024, our platform manages over $24bn in complex credit assets, standardizing over +35bn data points coming from 130+ systems all over the world.

Ready to Redefine Work? Join Cardo AI and let boldness and creativity reign. Together, we’ll shape the future of fintech! Apply now.