Summary
Vadim Kirpa is a Lead Data Scientist with 11 years of experience building and scaling ML solutions for credit products, currently leading data science at Wildberries after a long impact-driven tenure at Sberbank. He has deep hands-on experience designing resilient credit scoring stacks, calibration pipelines, uplift and graph-based risk models, and has led teams of 3–7 data scientists while also owning production code, audit interactions and financial impact evaluation. Early success in ICPC and a master’s in Mathematics and Computer Science underpin a strong algorithmic and analytical foundation that he applies to difficult ETL, model approximation (including a 1000x speedup), and AutoML work. He balances rigorous, regulator-friendly modeling (explainable income models) with experimental R&D—exploring integral/meta-models that consume other model scores and robustness to ill or missing domains. Outside of work he studies Japanese (JLPT N4), plays piano and experiments with Rust, reflecting both disciplined learning and low-level systems curiosity.
11 years of coding experience
5 years of employment as a software developer
Master's degree Mathematics and Computer Science, Master's degree Mathematics and Computer Science at Far Eastern Federal University
Japanese, English, Russian