Sergei Mironov
Quantum Software Developer at Entropica Labs
Yerevan, Armenia
Summary
🤩
Rockstar🎓
Top SchoolSergei Mironov is a Quantum Software Developer with 13 years of systems- and compiler-focused engineering experience, now building fault-tolerant quantum computation tooling at Entropica Labs from Yerevan. He has led compiler and ML infrastructure projects at companies including Huawei, Hisilicon and Xanadu, where he maintained a JAX frontend for a quantum compiler and extended MLIR-based compilation subsystems. Sergei’s background spans low-level systems work (SoC bring-up, bootloaders, DPDK) to high-level ML compiler contributions—he added argmax/argmin and improved matmul in the widely used Apache TVM and hardened tokenization and pretraining pipelines in TensorFlow models. Comfortable designing domain-specific compilers and integrating active quantum error correction into hardware-aware toolchains, he favors roles that leverage strong mathematical foundations. An open-source pragmatist, he combines production-grade engineering with research-facing innovation and a track record of turning prototypes (TASO, tensor-network engines) into platform-integrated components.
13 years of coding experience
17 years of employment as a software developer
Master of Technology - MTech, Electrical, Electronic and Communications Engineering Technology/Technician, Master of Technology - MTech, Electrical, Electronic and Communications Engineering Technology/Technician at Bauman Moscow State Technical University
English, Russian
Stackoverflow
926reputation
146kreached
18answers
7questions
Github Skills (20)
tvm10
tokenize10
python10
machine-learning10
data-preprocessing10
deeplearning-ai10
tokenizer10
compiler-compiler10
deep-learning10
tensorflow10
bert10
compiler10
nlp10
bleu9
haskell6
Programming languages (25)
Github contributions (5)
apache/tvm
Jul 2018 - Dec 2018
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
ML Engineer
ML EngineerContributions:8 commits, 8 PRs, 89 comments in 4 months
Contributions summary:Sergei contributed to the implementation of machine learning operations within the TVM compiler stack. They added argmax and argmin operations, which are essential for tasks like finding the most probable class in a classification model. Their work included writing tests and integrating these operations into the TensorFlow frontend, demonstrating a focus on extending TVM's capabilities in support of deep learning models. The user also made code adjustments to improve the functionality of the matmul function.
metalvulkancompilertensoropencl
tensorflow/models
Mar 2020 - Apr 2020
Models and examples built with TensorFlow
Role in this project:ML Engineer
ML EngineerContributions:6 commits, 10 PRs, 12 comments in 1 month
Contributions summary:Sergei primarily focused on modifying the tokenizer component within the TensorFlow models repository. Their contributions involved enhancements to the tokenizer's functionality, including safety checks and un-hardcoding of character sets. They also made modifications to the BERT pretrainer and data processing scripts, demonstrating involvement in model training and related functionalities, which also included making output numbers of unmatching lines in BLEU. This suggests a focus on improving the pretraining and tokenization processes within the context of NLP models.
deep-learningtensorflow
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.