Skip to content
View mprahboamey's full-sized avatar
🧑‍🍳
🧑‍🍳

Block or report mprahboamey

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
mprahboamey/README.md

Hi there 👋, I'm Mprah-Boamey

Independent builder · optical neural network simulation · memory-mapped weight runtimes · Ghana

Profile views


💫 About Me

  • 🔭 Now: Optical neural network simulation in PyTorch; crossing from simulator into hardware. Pinned repos are the main work.
  • 👯 Open to: Optical computing research, inference-efficiency collaborations, photonics or unconventional-hardware groups
  • 🌱 Learning: Holographic storage materials, photonic device physics, hypergraph models of discrete physics
  • 💬 Ask me about: PyTorch wave optics, the Angular Spectrum Method, memory-mapped tile runtimes, GGUF internals
  • Fun fact: Learned linear algebra by trying to implement self-attention as wave interference
  • 🌐 Portfolio: mprah-boamey.vercel.app

🌐 Socials

LinkedIn X GitHub


💻 Tech stack

Python TypeScript JavaScript C Bash HTML5 CSS3 Markdown YAML LaTeX Node.js Express FastAPI React Next.js TailwindCSS PyTorch TensorFlow Keras NumPy SciPy Pandas Scikit-learn Matplotlib Plotly MLflow Jupyter CUDA Ollama GGUF PyBullet Webots Linux Git GitHub GitLab Docker NPM Vercel AWS SQLite PostgreSQL MySQL Redis Prisma Supabase JSON VS Code Cursor


📊 GitHub stats

GitHub stats GitHub streak Top languages

Self-taught · 19 · Ghana

Pinned Loading

  1. PHOTEX-clean PHOTEX-clean Public

    PyTorch simulator for optical neural networks. Trainable diffractive layers, Angular Spectrum Method wave propagation, wave-interference self-attention. Benchmark tooling vs published NVIDIA H100 f…

    2

  2. holoweights holoweights Public

    Memory-mapped 128×128 uint8 tile layout for neural network weights with sparse routing. Software-only; measured RSS savings against dense FP16 baselines on GGUF artifacts.

    Python 2