Skip to content
View Mattral's full-sized avatar
👀
I may be slow to respond.
👀
I may be slow to respond.

Block or report Mattral

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
mattral/README.md

🌌 Mattral

Building systems that see and understand

ML Engineer · Computer Vision · NLP · Part-time human being


mattral


✨ Between the lines

I build machine learning systems that bridge the gap between research and reality.
Specializing in computer vision and language models, I care about things that work—and work at scale.

My work lives in two worlds:
the clean abstractions of mathematics,
and the messy reality of production systems.

Current state of affairs:
Building proprietary systems by day,
exploring open questions by night.
(GitHub activity is the calm between storms)


Some truths, in no particular order:

  • Morning ritual: coffee, ambient music, checking if the model converged
  • Believe in elegant systems over clever hacks
  • Forever fascinated by how machines learn to see
  • Existential crisis: optional but often included
  • Keep my life private, my code public, and my commits clean
  • Currently more active in proprietary realms than GitHub's public squares

🛠 Tools of the trade

Selected with intention, used with purpose.

Perception & Intelligence


Infrastructure & Scale


Data & Systems


Craftsmanship



🌐 Connect & context



🎶 Current frequency



📊 Rhythm & motion

The quiet work behind the scenes.

mattral

mattral


snek

contribution-graph-3d
contribution-graph
activity-graph
banner

Where mathematics meets the machine,
and curiosity meets code.


"Most of my interesting work happens in private repositories.
Consider this GitHub the quiet cafe where I work on side projects
when the proprietary systems are sleeping."

Pinned Loading

  1. ML-AI-Algorithms-from-scratch ML-AI-Algorithms-from-scratch Public

    implementation of several Supervised, Unsupervised, Bayesian, Neural Networks and Reinforcement Learning Algorithms from scratch with only numpy and basic libs.

    Python 33 3

  2. RAG-Multimodal-Financial-Document-Analysis-and-Recall RAG-Multimodal-Financial-Document-Analysis-and-Recall Public

    (RAG) Developing a chatbot capable of utilizing PDF files as a knowledge base to answer questions. Additionally, we employed the vision capability of GPT-4V to identify and describe graphs from eac…

    Jupyter Notebook 50 11

  3. NOTE-Best-Practices-for-Computer-Vision NOTE-Best-Practices-for-Computer-Vision Public

    Best Practices for Computer Vision

    Python 44

  4. Kolmogorov-Arnold-Networks Kolmogorov-Arnold-Networks Public

    customizable neural network architecture based on Kolmogorov-Arnold Networks (KANs), utilizing TensorFlow

    Python 19 1

  5. Click-Analytics Click-Analytics Public

    Do what Analysts usually do with just mouse clicks, Unstable Yet, but can tryout deployed version. Referenced and modified from an older repository.

    Python 7 1

  6. LLM-Improving-Trained-Models-with-RLHF LLM-Improving-Trained-Models-with-RLHF Public

    Experimented with the three essential Reinforcement Learning with Human Feedback (RLHF) process stages. It starts by revisiting the Supervised Fine-Tuning (SFT) process, then proceeds with the trai…

    Jupyter Notebook 13 2