Skip to content
View kerimcaliskan182's full-sized avatar
5 followers · 6 following

Achievements

Achievement: Starstruck

Achievements

Achievement: Starstruck

Block or report kerimcaliskan182

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
kerimcaliskan182/README.md

About Me

I am a Physicist and Embedded Software Engineer with expertise in Computational Physics, Embedded Systems, Physics-Informed Neural Networks (PINNs), and Machine Learning. I hold a degree in Physics from Middle East Technical University (METU) and currently work as a full-time Software and R&D Engineer at TDG (Teknik Destek Grubu).

Embedded Systems Projects

1. Embedded Systems Experience with Linux and MCU Programming

I have worked extensively with Linux OS and various communication protocols such as UDP and TCP. I have also written firmware for microcontroller units (MCUs), handling communication, real-time data processing, and low-level hardware interactions.

Key Contributions:

  • Developed and optimized communication protocols using UDP/TCP for embedded systems to ensure efficient real-time data transfer.
  • Wrote and tested firmware for MCUs, focusing on low-level hardware programming and real-time system requirements.
  • Worked in Linux environments to manage, debug, and optimize embedded system applications.

2. Firmware Development

I developed firmware for sensors using Mbed OS 6, focusing on reliable and efficient operation. This involved embedded programming in C++ and optimizing sensor performance for real-time applications.

Key Contributions:

  • Designed and implemented firmware to enhance sensor data collection and processing.
  • Improved system reliability and responsiveness through embedded software optimization.
  • Collaborated with hardware teams to integrate software solutions into the sensor platform.

Deep Learning & Computational Physics Projects

1. Solving the Schrödinger Equation Using Deep Learning

This project solves the 3D Schrödinger Equation using a Physics-Informed Neural Network (PINN). The model predicts electron probability density and energy eigenvalues with high accuracy.

  • Tools & Technologies: Python, PyTorch, SciPy, Matplotlib
  • Key Features:
    • Solved the equation in spherical coordinates using a neural network.
    • Achieved energy eigenvalue close to the theoretical value (-13.59 eV).
    • Visualized the electron probability density.

2. Solving the Helmholtz Equation Using Deep Learning

This project applies a deep learning approach to solve the Helmholtz Equation using PINNs and boundary condition transformations.

  • Tools & Technologies: Python, PyTorch, SciPy, Matplotlib
  • Key Features:
    • Implemented sampling techniques like Quasi-Monte Carlo and Latin Hypercube Sampling.
    • Visualized true vs. predicted solutions and error analysis.

Helmholtz Equation Solution using PINN

3. Fluid Dynamics Simulation Using Navier-Stokes Equations

This project simulates fluid dynamics using the Navier-Stokes Equations for a cylinder flow model.

  • Tools & Technologies: Python, PyTorch, Matplotlib, NumPy
  • Key Features:
    • Used PINNs to model velocity and pressure fields over time.
    • Validated results against experimental data.
    • Handled complex simulation datasets with precision.

Navier-Stokes Equation Solution using PINN

Technical Skills

  • Programming Languages: C++, Python, MATLAB, LATEX
  • Frameworks & Libraries: PyTorch, TensorFlow, SciPy, Matplotlib, NumPy
  • Embedded Systems: Mbed OS, Linux, UDP/TCP, MCU programming, Firmware development
  • Tools: Linux, Git, MS Office Programs
  • Specializations: Embedded Systems, Deep Learning, Physics-Informed Neural Networks, Computational Physics, Machine Learning

Contact

Feel free to reach out to me on LinkedIn or check out my GitHub profile for more projects.

Pinned Loading

  1. NavierStokesEqSolnWithPINN NavierStokesEqSolnWithPINN Public

    This Python script solves the Navier-Stokes equations using Physics-Informed Neural Network. This approach enables the modeling of fluid dynamics problems by learning the velocity field and pressur…

    Python 16

  2. Helmholtz-eq-soln-PINN Helmholtz-eq-soln-PINN Public

    Python 1