Dmitry Sviridkin

Contact Profile

Software engineer, passionate with high performance low-level software development and software safety. Confident with multiple programming languages and ready to adapt new tools.

Skills

Programming languages: Rust, C++, Python, ASM

Linux, system level development, algorithms and data structures, profiling, benchmarking and optimization, statistics and machine learning basics, mentoring and code reviews.

Education

Master of Computer Science

Southern Federal University, Rostov-on-Don, Russia (2017-2019)

Master's dissertation researched problems of optimal creations leaks in oriented transport networks.

Bachelor of Applied Math

Southern Federal University, Rostov-on-Don, Russia (2013-2017)

Work Experience

Software Engineer II - Amazon Web Services, CloudFront

October 2022 - Current

• Developed Cloudfront KVS feature backend with sub millisecond latency.
• Designed and conducted code quality & safety reviews for CloudFront Functions Origin Selection Feature.
• Led the project for optimizations of Cloudfront Functions.
• Rolled out Cloudfront JavaScript engine optimizations.
• Identified and fixed various memory-related vulnerabilities.
• Maintained a secure sandboxed environment for JavaScript code via low-level Linux facilities (seccomp, eBPF, cgroups, memory hardening).
• Mentored team members for onboarding Rust and delivering safe Rust - C/C++ FFI layers.

Senior Software Engineer

Arrival July 2021 - September 2022

• Integrated computer vision subsystem with a KUKA Robot controller.
• Developed hand-to-eye calibration algorithms for a 3D vision sensor.
• Created feature detection algorithms on 3D point clouds using OpenCV and CUDA.

Visiting Lecturer - Saint-Petersburg State University

February 2021 - May 2022

• Developed full Linux programming course materials including presentations, assignments, and exams.
• Supervised and mentored students on various projects.

Software Engineer - Speech Technology Center

August 2019 - June 2021

• Optimized memory and speed for a hybrid speech recognition decoder to meet real-time requirements with sub-second latency.
• Developed keyword search algorithms for decoded word lattices.

Visiting Lecturer - Higher School of Economics

September 2019 - May 2020

• Conducted practice classes for a C++ programming course.

Software Engineer - Research Institute of Radio Communications

July 2017 - July 2019

• Worked on digital signal processing algorithms for embedded devices.
• Developed UIs with the Qt framework.
• Built a Linux kernel module with DMA for high bandwidth receivers.

Publications & Projects

Author of the UBBook — the guide for interactions between Undefined Behavior and C++ compiler optimizations. https://github.com/Nekrolm/ubbook

Collection of tricky notes with C++ examples. https://gist.github.com/nekrolm

Languages

Russian, English