Skip to content
View JayThibs's full-sized avatar
:atom:
Always building.
:atom:
Always building.

Block or report JayThibs

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
JayThibs/README.md

Jacques Thibodeau

Typing SVG

Blog  X  LinkedIn


AI safety researcher turned founder. I spent years deep in alignment work, from SERI MATS to leading the Supervising AIs Improving AIs research agenda to building tools that accelerate safety researchers. Previously a data scientist and physicist.


Writing

I blog at Aligned to Flourish about AI safety, building in the age of agents, and the future of software.

Recent

    When Execution Gets Cheap, Does Taste Become the Moat?
    Hard Truths About Where AI Is Headed
    Difficulties in Building an AI Safety Startup
    Gaining Clarity on Automated Alignment Research
    Better Model Diffing Is Needed
    Automating AI Safety: What We Can Do Today

AI Alignment

    AI Alignment Project Ideas
    Accelerating AI Alignment Research (Talk)
    Using Data Attribution for AI Alignment
    A Descriptive Overview of Current AI Alignment Research

Miscellaneous

    How Much I'm Paying for AI Productivity Software
    The Importance of Entropy
    Quantum Computing, Photonics, and Energy Bottlenecks for AGI

All posts →


GitHub Stats

GitHub Streak

Pinned Loading

  1. moirage/alignment-research-dataset moirage/alignment-research-dataset Public

    A dataset of alignment research and code to reproduce it

    HTML 78 18

  2. supervising-ais-improving-ais supervising-ais-improving-ais Public

    Future prosaic AIs will likely shape their own development or that of successor AIs. We're trying to make sure they don't go insane.

    Python 4 1

  3. Weak-Supervised-Learning-Case-Study Weak-Supervised-Learning-Case-Study Public

    Exploring NLP weak supervision approaches to train text classification models. The project is also a prototype for a semi-automated text data labelling platform. Approaches: Snorkel and Zero-Shot L…

    Jupyter Notebook 22 4

  4. map-floodwater-satellite-imagery map-floodwater-satellite-imagery Public

    This repository focuses on training semantic segmentation models to predict the presence of floodwater for disaster prevention. Models were trained using SageMaker and Colab.

    Jupyter Notebook 33 8

  5. CER-REC/esa-data-bank_banque-donnees-ees CER-REC/esa-data-bank_banque-donnees-ees Public

    BERDI's environmental and socio-economic assessment (ESA) data extraction method | Notre méthode d'extraction des données pour les évaluations environnementales et socioéconomiques (« EES ») de CIBER

    Python 7 1

  6. hyperdrive-vs-automl-plus-deployment hyperdrive-vs-automl-plus-deployment Public

    Comparing models trained with HyperDrive and AutoML, and choosing the best model to deploy and consume in AzureML Studio.

    Jupyter Notebook 1