Skip to content
View Mechres's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report Mechres

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
Mechres/README.md

Hi, I'm Yağız 👋

MIS student · systems tinkerer · hardware hacker
I build things from the bootloader up - JIT compilers, custom Linux daemons, and boards I probably shouldn't have opened.


⚡ Currently

  • Optimizing RandomX's AArch64 JIT compiler (armrx) - chasing down branch mispredictions and closing an instruction-count gap against XMRig
  • Building Salca, a from-scratch C++/OpenGL isometric action slasher roguelike
  • Studying Management Information Systems @ Erciyes University, with coursework spanning strategic management, data analysis, and ML

🔩 Also known for

  • Hardware & Firmware: Board-level electronics repair, firmware reverse engineering, and giving dead hardware a second life - BIOS chip-level fixes, NAND recovery, embedded IR/RF protocol implementations, and a homebrew 8088 retrocomputer built from salvaged ICs.
  • Embedded Linux & Drivers: Bringing up custom operating systems (postmarketOS, antiX) on non-standard/legacy mobile and tablet platforms, and writing custom Linux kernel drivers/systemd daemons to remap and control proprietary hardware.
  • Automotive Hacking: Designing and building custom transistor-based level-shifter circuits to interface with and read K-Line diagnostic data directly from vehicle ECUs.

🌐 Homelab & Infra

  • Self-hosted services behind Cloudflare Zero Trust tunnels
  • Multi-node Ubuntu setup with failover experiments
  • Local AI inference (llama.cpp, Vulkan / ROCm benchmarking)
  • DIY monitoring, automation, and service orchestration

🤖 Data & ML

  • Feature engineering + model stacking (XGBoost, CatBoost, RF)
  • AutoML experimentation (TPOT) and optimization pipelines
  • Kaggle-style datasets and prediction systems
  • Local LLM experimentation (quantization, GPU backends, performance tuning)

🛠️ Tech I reach for

📌 Pinned work


📊 Stats


Always down to talk JIT compilers, board-level repair, or roguelike architecture - open an issue or PR anytime.

Pinned Loading

  1. Price-Predict Price-Predict Public

    Crypto & Stock* price prediction with regression models.

    Python 14 4

  2. song-recommender song-recommender Public

    This Flask-based application utilizes Spotify's API and machine learning techniques to provide personalized song recommendations. It analyzes track features (like tempo, danceability, energy) and g…

    Python 2

  3. Pareidolia-attention-test Pareidolia-attention-test Public

    Pareidolia attention test

    Python 2

  4. YouSpoPls YouSpoPls Public

    Simple & effective, Youtube - Spotify, Spotify - Youtube playlist copier.

    Python 4

  5. text-summarize text-summarize Public

    Flask-based API that provides a user-friendly interface to summarize text in any language supported by the BART model and then translate it to English.

    Python 2 1

  6. logitech-g710-linux-driver-daemon logitech-g710-linux-driver-daemon Public

    Forked from Wattos/logitech-g710-linux-driver

    Logitech G710 Linux Driver

    C 1