Project title: Data-driven approaches towards the optimization of PbTe-based thermoelectric materials

Description:
developing machine-learning pipelines for structure–property prediction of doped PbTe-based thermoelectric materials by applying LoRA-based low-rank adaptation techniques to materials-science datasets. Analyzing dopant-induced effects on lattice dynamics and brittle behavior using data-driven models, and assisting in computational and ML-driven workflows relevant to thermoelectric materials characterization. Results of simulations can be included in the publication in the journal (if successful)

Candidate requirements:
➔ knowledge of programming in python
➔ skills in machine learning
➔ ability to work in a team
 ➔ basic knowledge in materials science and atomistic simulations 
➔ desire to learn and work

Supervisor and contact person: Alexander Kvashnin, a.kvashnin@skoltech.ru

Internship duration: from 1 month

Deadline for submission: February 15, 2026

The start date of the internship: March 2026

Compensation: not available