Skip to content
View nschaetti's full-sized avatar
🎯
Focusing
🎯
Focusing
100 followers · 23 following

Achievements

Achievement: Arctic Code Vault ContributorAchievement: Starstruckx2

Achievements

Achievement: Arctic Code Vault ContributorAchievement: Starstruckx2

Block or report nschaetti

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
nschaetti/README.md

I'm a Swiss researcher in Machine Learning and Artificial Intelligence, working at the intersection of dynamical systems, reservoir computing, self-supervised learning, and simulation-based inference.

My work combines theoretical and applied machine learning, scientific computing, and visual tools for research and education. I focus on models that are understandable, controllable, and grounded in physical or dynamical principles, with a strong commitment to free and open-source AI (FOSAI & FOSS).

Research interests

  • Reservoir Computing (RC) & Echo State Networks (ESN)
  • Self-Supervised Learning (SSL): contrastive and regularized learning (JEPA)
  • Physics-informed, neural operators for simulation-based inference (SBI)
  • Visual computing & generative effects for arts and education

Selected projects

  • Artificialis: a YouTube channel exploring artificial intelligence through code, mathematics, and critical thinking, without ignoring its philosophical, artistic, and social dimensions.
  • EchoTorch: a Python toolkit for Reservoir Computing and Echo State Network experimentation built on top of PyTorch.
  • Pixel Prism: a Python library for mathematical education with advanced visual effects and symbolic mathematics.
  • Hyperion: an educational deep learning stack aimed at demystifying modern ML systems.
  • DeckPilot – customizable Stream Deck interface for scripting and automation
  • Tempest – CUDA-accelerated engine for full waveform inversion and scientific modeling
  • boardGPT – lightweight experiments around training/fine-tuning GPT-like models on structured games and languages for AGI.

Background

  • PostDoc Researcher in Machine Learning & AI
  • Experience across academia, applied research, and open-source development
  • Strong background in Python, C++, CUDA, and scientific ML tooling

Links

Pinned Loading

  1. EchoTorch EchoTorch Public

    A Python toolkit for Reservoir Computing and Echo State Network experimentation based on pyTorch. EchoTorch is the only Python module available to easily create Deep Reservoir Computing models.

    Python 489 116

  2. PixelPrism PixelPrism Public

    PixelPrism is a Python library that applies advanced visual effects to videos and images

    Python

  3. DeckPilot DeckPilot Public

    DeckPilot is a customizable interface for your Stream Deck, allowing you to assign scripts to buttons, manage multiple windows, and navigate through hierarchical categories. Designed for flexibilit…

    Python

  4. boardGPT boardGPT Public

    The simplest, fastest repository for training/finetuning GPTs on board games

    Jupyter Notebook 1

  5. Tempest Tempest Public

    A CUDA-accelerated engine for Full Waveform Inversion and seismic forward modeling

    C++ 1

  6. Hyperion Hyperion Public

    Hyperion is a full educational deep learning stack designed to reach expert-level understanding of modern machine learning systems, from tensor computation to Transformer architectures and GPU kern…