About
I am currently a Senior Data Scientist at Danske Bank and hold a PhD from the Institute of Data Science and Digital Technologies at Vilnius University. I have more than 6 years of developer experience. Previously, I spent 4 years in London working in various engineering positions at JP Morgan, UBS, and CME Group.
Education
| Year | Degree | Institution |
|---|---|---|
| 2019-2024 | Informatics, Ph.D. | Vilnius University, Lithuania |
| 2014-2017 | Informatics, MS | Vilnius University, Lithuania |
| 2007-2011 | Economics, BS | Kaunas University of Technology, Lithuania |
Publications
- Chang, T. A., et al. Global PIQA: Evaluating Physical Commonsense Reasoning Across 100+ Languages and Cultures. arXiv preprint arXiv:2510.24081 (2025). (Contributor)
- Tautvaišas, S., Žilinskas, J. Heteroscedastic Bayesian Optimization using Generalized Product of Experts. Journal of Global Optimization (2023). Link
- Tautvaišas, S., Žilinskas, J. Scalable Bayesian optimization with generalized product of experts. Journal of Global Optimization (2022). Link
Work experience
Danske Bank - Vilnius, Lithuania
Senior Data Scientist, GenAI
02/2025 – current
- Researching and evaluating AI models for innovative AI products;
- Creating benchmarks and evaluation for AI models, RAG systems, and agents;
- Advising business domains on adopting generative AI solutions;
- Presenting latest AI research in internal reading group.
Senior Data Scientist, SuperAI
07/2021 – 02/2025
- Development and maintenance of credit risk models (probability of default);
- Worked on data quality and feature engineering for new SME risk model;
- Building tools to improve the monitoring of customers credit scores;
- Ad-hoc data analysis for Danish financial regulators;
- Redeveloped capital impact model from R to Python;
- Created PoC for new default collection using DBT library;
- Proposed and implemented new credit rating override thresholds methodology;
- Technology stack: Python, R, STATISTICA, SQL, Azure DevOps, Git.
JP Morgan Asset Management - London, United Kingdom
Software Engineer, Multi-Asset Solutions Technology
08/2018 – 06/2019
- Building new generation investment platform for multi-asset solutions portfolio managers;
- Developing and improving benchmark service application to compare portfolios against benchmarks;
- Created new application in ReactJS to visualize FX rates historical data;
- Technology stack: Java 8, C#, Spring Boot, ReactJS, HighchartsJS.
UBS Investment Bank - London, United Kingdom
Java Developer, Rates & Credit Market Technology
04/2017 – 07/2018
- Worked on developing low latency Smart Order Router based on Aeron messaging system and binary encoding;
- Developed performance tests for SOR and different messaging queues (TIBCO, Disruptor, Aeron);
- MIFID II development duties for SOR and Fix Gateways;
- Joined new cross-functional team to develop Algo trading platform for FX and Rates based on low latency Equities Algo Execution Services platform;
- Technology stack: Java 8, ION MarketView, JMH, Disruptor, RxJava, KDB, FIX.
CME GROUP - London, United Kingdom
Software Engineer, Risk and Pricing Systems Department
12/2015 – 04/2017
- Developed historical scenarios reference data module working with Quants for Futures/Options/OTC products P&L and Risk pricing;
- Development duties for Admin UI (JavaScript, AngularJS, MongoDB) for supporting P&L and Risk pricing application;
- Created Cucumber automated tests for risk pricing engine;
- Developed python scripts to compare different risk models (HVar, SPAN) using client portfolios and historical market data;
- Technology stack: Java 8, Python, Angular JS, Spring Boot, Mongo DB, Karma, Jasmine, Cucumber, JUnit, Splunk.
CF Partners Asset Management LLP - London, United Kingdom
Intern Quantitative Python Developer
09/2015 – 12/2015
- Internship at a London-based specialist Long/Short commodities/utilities hedge fund;
- Created an automated stock screener and backtester using Python, Numpy, Pandas, and Bloomberg API;
- Researched financial ratios importance for screener performance and implemented GA algorithm for optimization;
- Researched causality relationships between commodities and stocks using VAR model.
Barclays Technology Centre Lithuania - Vilnius, Lithuania
Support Analyst
12/2013 – 08/2015
- Level 3 support for client-facing personal banking application;
- Fixing Java and JavaScript code defects, implementing application improvements;
- Technology stack: Java 6, JavaScript, JQuery, CSS, Spring MVC, MySQL.