markotesic.org
markotesic.org
Home
CV
Tutorials
Publications
Past Projects
Contact
Projects
Mistral AI Hackathon - CompanionAI
I developed a conversational companion for elderly individuals and those with memory challenges. The companion, implemented as a Telegram bot and using Mistral AI LLMs as backend, maintains conversational history and is designed to be empathetic toward the user.
Marko Tešić
,
Alex Chalakov
GitHub
Devpost
Workshop on Human Behavioral Aspects of (X)AI
I organised a workshop that brought together researchers from machine learning and cognitive science to discuss the behavioral aspects of explainable AI.
(Un)interesting correlations: What are the chances that correlations lead to causation?
We use directed acyclic graphs (DAGs) to investigate the chances that two variables are causally connected, correlated, and that a covariate is inducing a correlation when controlled for.
Marko Tešić
,
Ulrike Hahn
,
Kirsty Phillips
,
Jason Burton
GitHub
Turing Data Study Group: Optimising the supply chain to minimise waste and delivery mileage
I worked on predicting deliveries to stores such that waste is minimised.
PDF
GitHub
(Causal) Bayesian modeling of investment factors and Environmental, Social and Governance (ESG) criteria
As part of the BlackRock’s Factor Based Strategies Group I worked on understanding how some ESG criteria such as carbon emissions can impact return on equity.
Cite
×