Skip to content
View yaroslavaristov's full-sized avatar
🎯
get sh#t done.
🎯
get sh#t done.
  • Amsterdam

Highlights

  • Pro

Block or report yaroslavaristov

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
yaroslavaristov/README.md

Yaroslav Aristov

C++ Software Engineer · Ultra Low-Latency · HFT Infrastructure


I engineer ultra low-latency trading systems in C++20/23 on Linux. Every design decision is driven by one constraint: latency. No exceptions. No virtual dispatch. No heap allocation in the critical path.


Engineering Focus

  • Lock-free concurrency — SPSC/MPMC ring buffers, cache-line padding against false sharing, wait-free progress guarantees
  • CPU-level optimization — cache-line aware data layout, SIMD/AVX vectorization, branch prediction control, prefetching
  • Kernel bypass — DPDK, RDMA, onload/ef_vi for sub-microsecond network I/O
  • Memory discipline — pool allocators, slab allocation, zero dynamic allocation in hot path, NUMA-aware design
  • Performance engineering — perf, Valgrind, flamegraphs, hardware PMU counters, L1/L2/LLC miss analysis

Stack

C++ Linux CMake Python Git GitHub Actions


GitHub Activity

GitHub Streak


Research

Acceleration of Parallel SAT Solver for Cryptanalysis of Hash Functions
Co-author · NSU Cryptography Summer School-Conference 2025 · pp. 132-142

Up to 41% runtime reduction on multi-threaded configurations via low-level algorithmic optimization.


Awards

WorldSkills — 2nd place, Neural Interface Design (2026)


Telegram ProtonMail


English — fluent. Open to relocation globally.

Pinned Loading

  1. quanttok quanttok Public

    Zero-alloc, SIMD-accelerated, compile-time-optimized BPE tokenizer. Faster than tiktoken, built for light speed tokenizing.

    CMake

  2. arena-allocator arena-allocator Public

    Zero-allocation bump-pointer arena allocator for HFT critical paths. mmap-backed, HugePage support, O(1) allocation, 10-342x faster than malloc.

    C++