Dimitri Persiianov
Founder & CEO at Extruct AI
Barcelona, Catalonia, Spain
Summary
👤
Senior🎓
Top SchoolDimitri Persiianov is a founder and CEO with 11 years of ML and data engineering experience, currently building Extruct AI, a B2B search engine, from Barcelona. He blends hands-on model and infrastructure work with product-led leadership—previously leading Search Ranking & Personalization at Constructor where he shipped real-time ML ranking, MLOps, and experimentation platforms. A pragmatic engineer who co-founded and scaled Condukt to a pre-seed-backed API product, he also has deep research-and-production pedigree from Yandex’s dialogue systems and a speed-up contribution to the Gensim library used widely in NLP. He mentors and hires engineering teams, drives system design and product discovery, and keeps code-level fluency in embeddings, Doc2Vec/Word2Vec/FastText, and ranking infra. Dimitri holds degrees from MIPT and YSDA, and his background combines academic rigor with measurable production impact—including large training optimizations that accelerated popular NLP tooling.
11 years of coding experience
6 years of employment as a software developer
Master's degree Computer Science, Master's degree Computer Science at Moscow Institute of Physics and Technology (State University) (MIPT)Master's degree Data Science, Master's degree Data Science at Yandex School of Data Analysis
English, Russian, Spanish
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (14)
word-embeddings10
gensim10
machine-learning10
nlp10
python10
natural-language-processing10
numpy9
testing9
word2vec9
unit-test9
topic-modeling9
unit-testing9
fasttext8
data-science8
Programming languages (7)
Github contributions (5)
piskvorky/gensim
Mar 2018 - Jan 2020
Topic Modelling for Humans
Role in this project:
Back-end Developer / ML Engineer
Back-end Developer / ML EngineerContributions:5 commits, 7 PRs, 140 comments in 1 year 9 months
Contributions summary:Dimitri primarily contributed to the `gensim` library, focusing on enhancing the `BaseKeyedVectors` class with new methods like `add_entity` and `__setitem__` to manually fill and manipulate word embeddings. They also worked on the multistream API for vocabulary building, integrating support for Doc2Vec, Word2Vec, and FastText models, including writing tests and addressing code style issues. Furthermore, the user addressed a critical overflow error in the file-based training implementation.
pythonword-similarityword-embeddingsdata-miningfor-humans
persiyanov/skip-thought-tf
Sep 2016 - Mar 2017
An implementation of skip-thought vectors in Tensorflow
Contributions:4 commits, 2 PRs, 40 pushes in 6 months
nlpskip-thought-vectorsdeep-learningembeddingsskip
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.