GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Sign in
  • ActionsCI/CD and automation
  • CodespacesCloud development environments
  • All Features
  • Documentation
  • Training
  • Certifications
  • Build with Copilot
  • Customize Copilot
  • Integrations
  • Trending repos
  • GitHub Sponsors
  • Developer articles
  • Enterprises

This view

  • Original page
  • New address

GeezerVison is showing github.com — jknoll - Overview

@jknoll jknoll Follow
More
  • Overview
  • Repositories
  • Projects
  • Packages
  • Stars
@jknoll
jknoll
Follow
View jknoll's full-sized avatar

Justin F. Knoll jknoll

Follow
Distributed systems, peer-to-peer, blockchain, Web3, ML, AI, Deep Learning, LLMs.
24 followers · 16 following
  • @bittorrent
  • San Francisco, CA
  • @jknoll
  • in/knolljustin

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx2Achievement: YOLOAchievement: QuickdrawAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx2Achievement: YOLOAchievement: QuickdrawAchievement: Arctic Code Vault Contributor
More
  • Overview
  • Repositories
  • Projects
  • Packages
  • Stars
jknoll/README.md

Hi, I'm Justin

I'm AI-native product manager and engineer focused on deep learning, LLM agents, and computer vision. I've won multiple hackathons and received research grants for work in training deep neural networks for chess and taxonomic classification.

AI project details jknoll.github.io | General Profile LinkedIn


Featured Projects

Chess Theme Classifier

Deep CNN for multi-label classification of chess positions. Identifies tactical themes (back rank mate, zugzwang, pins) and openings from board state. Trained on 3M+ lichess puzzles.

Grant Recipient

PyTorch Computer Vision Multi-label Classification

FungiTastic Classifier

Modified BEiT vision transformer for hierarchical taxonomic classification of fungi across six biological ranks. 89.91% accuracy at phylum level on 600K+ images.

$5K Research Grant

Vision Transformers Hierarchical Classification BEiT

AdFlow

AI-powered video ad generation from product URLs. Uses Claude Agent SDK for multi-agent orchestration, web scraping for metadata extraction, and video generation APIs.

Agentic Orchestration Hack

Claude Agent SDK Multi-Agent Video Generation

ClaudeCAD

Connects Claude Desktop to Blender via MCP for 3D model generation from USPTO technical drawings. Describe a mechanism, get a 3D model.

YC/Anthropic MCP Hackathon

MCP Blender 3D Modeling

Pinned Loading

  1. chess-theme-classifier chess-theme-classifier Public

    This project implements a deep convolutional neural network to perform multi-label classification on board positions sourced from the lichess puzzles dataset. Each board position is labeled with ap…

    Python

  2. ai-chess-bot ai-chess-bot Public

    Train a deep neural network to learn the board position eval() function from labeled lichess.com position examples and use the result to play live chess.

    Python

  3. claude-cad claude-cad Public

    Connects Claude Desktop to a locally hosted Blender MCP server, a local Blender instance, and a newly-developed local MCP server which serves as a technical drawing database and server. The user ca…

    Python 3

  4. claude-plays-pokemon claude-plays-pokemon Public

    Forked from swyxio/morphcloud-examples-public

    Collaboration repo for Claude plays Pokemon

    Python 1

  5. mecanum-robot mecanum-robot Public

    Control system for simple holonomic drive mecanum-wheeled Raspberry Pi based robot.

    Python

  6. elo-ladder elo-ladder Public

    Super minimal no-frills code to calculate and display a minimal Elo Ladder for two player zero-sum games such as chess, billiards, etc. Previously used for continuous local billiards ladder.

    Python

Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.

Uh oh!

There was an error while loading. Please reload this page.

You are visitor 01402520 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.