Karim Knaebel

ViGeo. ViGeo: Towards Consistent Video Geometry Estimation

zed-101-setup. Your ultimate guide to setting up Zed with Vim mode, tailored settings, and key bindings for a seamless coding experience

vipe. ViPE: Video Pose Engine for Geometric 3D Perception

markdown-preview. A simple Markdown viewer for reading .md files

TeXicode. TeXicode - TeX to Unicode, a Unicode TeX renderer that renders TeX into Unicode by rendering the TeX

lgtm. Rust

lingbot-vision. Self-supervised learning for spatial perception

exercises-dataset. 1,324-exercise fitness dataset — animation GIFs, 180×180 thumbnails, muscle-group & equipment data, and step-by-step instructions in 6 languages. The exercise data layer behind the LogPress app.

CVIL. Practical CV/ML interview checklist used to land a computer vision internship. Phases + specialization tracks. Keep it CVIL.

lazyslurm. like lazygit/lazydocker but for slurm

Awesome-4D-Spatial-Intelligence. A curated list of awesome papers for reconstructing 4D spatial intelligence from video. (arXiv 2507.21045)

models.dev. An open-source database of AI models.

whisperX. WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)

jj-starship. Rust

pi. AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

codiff. a fast local diff viewer

helium. Private, fast, and honest web browser

trackio. A lightweight, local-first, and free experiment tracking library from Hugging Face 🤗

Conference-Acceptance-Rate. Acceptance rates for the major AI conferences

Open-d4rt. Python

CrossInk. Firmware for Xteink X3/X4. Personal fork of Crosspoint.

MDA. Modeling Depth Ambiguity: A Mixture-Density Representation for Flying-Point-Free Depth Estimation

surge. SurGe: Improved Surface Geometry in Point Maps

OpenLogi. ⚡️A native, local-first alternative to Logitech Options+, written in Rust 🦀 — remap buttons, DPI, and SmartShift over HID++. No account, no telemetry.

epub-to-xtc-converter. EPUB optimizer and XTC/XTCH converter for Xteink X4 e-reader (480×800 e-ink display)

crosspoint-reader. Firmware for the Xteink X3 and X4 e-readers

cmux. Open source Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents. Built for multitasking, organization, and programmability.

open_clip. An open source implementation of CLIP.

vggt-omega. [CVPR 2026 Oral] VGGT Omega

zed-ltex. LTeX+ Language Server support for Zed

Overleaf-Desktop. A native macOS app for syncing projects to your local filesystem

d4rt. Implementation of D4RT, Efficiently Reconstructing Dynamic Scenes, from Deepmind

clarity-template. Clarity: A Minimalist Website Template for AI Research

timme. timm, evolved

rtk. CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies

opentui. OpenTUI is a library for building terminal user interfaces (TUIs)

hunk. Review-first terminal diff viewer for agentic coders

ai-deadlines. :alarm_clock: AI conference deadline countdowns

Volt. Volume Transformer: Revisiting Vanilla Transformers for 3D Scene Understanding (ECCV 2026)

MuM. [CVPR26] MuM's a pretty good feature extractor for 3D tasks, probably the best.

lingbot-map. A feed-forward 3D foundation model for reconstructing scenes from streaming data

zed-latex. Rust

diskonaut. Terminal disk space navigator 🔭

LoMa. [ECCV 2026 Oral] LoMa: Local Feature Matching Revisited

eye-declare. A declarative inline TUI rendering library for Rust, built on Ratatui

TorchCode. 🔥 LeetCode for PyTorch — practice implementing softmax, attention, GPT-2 and more from scratch with instant auto-grading. Jupyter-based, self-hosted or try online.

iwe. Markdown knowledge graph — LSP for your editor, CLI + MCP memory for your AI agents

git-filter-repo. Quickly rewrite git repository history (filter-branch replacement)

giscus. A commenting system powered by GitHub Discussions. :octocat: :speech_balloon: :gem:

ZipMap. [CVPR 2026] ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training

cli. Google Workspace CLI — one command-line tool for Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more. Dynamically built from Google Discovery Service. Includes AI agent skills.

fonttools. A library to manipulate font files from Python.

flashoptim. Python

Lilex. 🤘Open source programming font

dotfiles. My personal dotfiles

pierre. pierre’s open source code

deff. interactive, side-by-side file review for git diffs with per-file navigation, vertical and horizontal scrolling, syntax highlighting, and added/deleted line tinting

research-monorepo. Monorepo contanining reusable model components, complete model architectures, and model training scripts.

zed-just. Zed extension for Justfiles

PyAV. Pythonic bindings for FFmpeg's libraries.

pipeline. Multithread python pipeline framework

fisher. A plugin manager for Fish

multi_gpu_training. Python

4RC. [ICML 2026] 4RC: 4D Reconstruction via Conditional Querying Anytime and Anywhere

clauntty. iOS Terminal with native Claude Code support

yoga. Yoga is an embeddable layout engine targeting web standards.

ogpk. a cli tool for observing og data

RayRoPE. [ECCV '26] Code for paper "RayRoPE: Projective Ray Positional Encoding for Multi-view Attention"

agent-stuff. These are commands I use with agents, mostly Claude

ImageCompare_vscode. TypeScript

zerobrew. A 5-20x faster experimental Homebrew alternative

wavelet-monodepth. [CVPR 2021] Monocular depth estimation using wavelets for efficiency

simpleeval. Simple Safe Sandboxed Extensible Expression Evaluator for Python

opencode. The open source coding agent.

InfiniDepth. [CVPR 2026] InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fields

hevi. ⚠️ MIGRATED TO CODEBERG ⚠️

Skills. My Codex Skills

DKT. [ICRA2026] official implement of "Diffusion Knows Transparency: Repurposing Video Diffusion for Transparent Object Depth and Normal Estimation"

representations4d. Jupyter Notebook

Pake. 🤱🏻 Turn any webpage into a desktop app with one command.

Mole. 🐹 Clean, uninstall, analyze, optimize, and monitor your Mac from the terminal.

NATTEN. Fast Multi-dimensional Sparse Attention

jjui. jjui is a TUI designed for interacting with the Jujutsu version control system.

fizzy. Kanban as it should be. Not as it has been.

radiance_meshes. Training code for "Radiance Meshes for Volumetric Reconstruction".

Lotus-2. Official implementation of "Lotus-2: Advancing Geometric Dense Prediction with Powerful Image Generative Model"

NAF. Official Implementation of NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering

zensical. A modern static site generator by the Material for MkDocs team

roma. RoMa: A lightweight library to deal with 3D rotations in PyTorch.

RoMaV2. Python

parqeye. Peek inside Parquet files right from your terminal

Depth-Anything-3. Depth Anything 3

itar. tar file index for constant-time member access

awesome-tips.

television. A very fast, portable and hackable fuzzy finder.

ListenerMusicPlayer. A Grace Material Design Music Player

BlenderProc. A procedural Blender pipeline for photorealistic training image generation

composition_rendering. Python

anyup. [ICLR '26 Oral] Official repository of the paper "AnyUp: Universal Feature Upsampling".

asimov. Automatically exclude development dependencies from Apple Time Machine backups

moderngl. Modern OpenGL binding for Python

RAE. Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

WarpConvNet. Make your wildest 3D ConvNet dream architectures come true

rift. a tiling window manager for macos

fastmod. A fast partial replacement for the codemod tool. Assists with large-scale codebase refactors via regex-based find and replace with human oversight and occasional intervention.

sws. Minimal, predictable, footgun-free config library.

IsaacLab. Unified framework for robot learning built on NVIDIA Isaac Sim

doggo. :dog: Command-line DNS Client for Humans. Written in Golang

choose. A human-friendly and fast alternative to cut (and sometimes awk)

modern-unix. A collection of modern/faster/saner alternatives to common unix commands.

TTT3R. [ICLR 2026] A simple state update rule to enhance length generalization for CUT3R

neural-visibility. Neural Visibility of Point Sets [SIGGRAPH Asia 2025]

mkdocs-material. Documentation that simply works

prek. ⚡ A fast Git hook manager written in Rust, designed as a drop-in alternative to pre-commit, reimagined.

once-campfire. Super simple group chat, without a subscription

UniCeption. Generalizable Perception Stack for all things 3D, 4D & Scene Understanding

map-anything. MapAnything: Universal Feed-Forward Metric 3D Reconstruction

sparse-vggt. [CVPR 2026] Block-Sparse Global Attention for Efficient Multi-View Geometry Transformers

slurmtop. Slurmtop is a refreshing and intuitive terminal user interface for monitoring SLURM clusters

homebrew-zathura. Homebrew formulae to build Zathura on Mac OS X

arxivit. LaTeX source cleaner with optional DPI-aware image resizing for arXiv uploads

vscode-helix. TypeScript

pypeln. Concurrent data pipelines in Python >>>

joblib. Computing with Python functions.

wBlock. The next-generation ad blocker for Safari.

clickbait-remover-for-youtube. Replaces thumbnails and modifies titles of videos on youtube.com

dinov3. Reference PyTorch implementation and models for DINOv3

googlesans-code. The Google Sans Code font family

KaTeX. Fast math typesetting for the web.

gpt-oss. gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

RePLAy-Release. Python

ai-deadlines. ⏰ AI conference deadline countdowns

torchtitan. A PyTorch native platform for training generative AI models

imagesize_py. Python

MINIMA. [CVPR 2025] MINIMA: Modality Invariant Image Matching

tinyio. Ever used asyncio and wished you hadn't? A tiny (~400 lines) event loop for Python.

Pi3. [ICLR 2026] π^3: Permutation-Equivariant Visual Geometry Learning

Franca. Official code of Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning

ratarmount. Access large archives as a filesystem efficiently, e.g., TAR, RAR, ZIP, GZ, BZ2, XZ, ZSTD archives

aistore. AIStore: scalable storage for AI applications

mpk. A command line interface that converts between JSON and MessagePack

Muon. Muon is an optimizer for hidden layers in neural networks

s5cmd. Parallel S3 and local filesystem execution tool.

PartCrafter. [NeurIPS 2025] PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers

awesome-domain-adaptation-and-generalization-for-3D. This is a collection of awesome works targeting domain adaptation and generalization for 3D data.

gemini-cli. An open-source AI agent that brings the power of Gemini directly into your terminal.

tarp. Fast and simple stream processing of files in tar files, useful for deep learning, big data, and many other applications.

DONUT. [ICCV 2025] DONUT: A Decoder-Only Model for Trajectory Prediction

cargo-edit. A utility for managing cargo dependencies from the command line.

watchfiles. Simple, modern and fast file watching and code reload for Python, written in Rust

icecream. 🍦 Never use print() to debug again.

webssl. Code for "Scaling Language-Free Visual Representation Learning" paper (Web-SSL).

Hunyuan3D-2.1. From Images to High-Fidelity 3D Assets with Production-Ready PBR Material

cuopt. GPU accelerated decision optimization

goose. an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

pueue. :stars: Manage your shell commands.

shellfn. Rust proc macro for easily and safely use shell scripts in rust

superansac. C++

focal-frequency-loss. [ICCV 2021] Focal Frequency Loss for Image Reconstruction and Synthesis

image. Encoding and decoding images in Rust

restic-automatic-backup-scheduler. Automatic restic backup using Backblaze B2 storage and either Linux systemd timers, macOS LaunchAgent, Windows ScheduledTask or simply cron.

easydict. Access dict values as attributes (works recursively)

faiss. A library for efficient similarity search and clustering of dense vectors.

tartanairpy. A Python package for the TartanAir-V2 dataset.

pyslam. pySLAM is a hybrid Python/C++ Visual SLAM pipeline supporting monocular, stereo, and RGB-D cameras. It provides a broad set of modern local and global feature extractors, multiple loop-closure strategies, a volumetric reconstruction module, integrated depth-prediction models, and semantic segmentation capabilities for enhanced scene understanding.

gloss. A light-weight Physically Based Renderer built in Rust

typed-ffmpeg. Type-safe FFmpeg bindings for Python & TypeScript — filters, typing, and docs

httpx. A next generation HTTP client for Python. 🦋

mlop. Next Generation Experimental Tracking for Machine Learning Operations

efficient. Utilities (mostly decorators) for efficient python, including concurrent, worker, rate limit, retry, disk cache.

spdl. Scalable and Performant Data Loading

florch. Flexible training scaffolding for PyTorch with various extensions and utils

zarr-python. An implementation of chunked, compressed, N-dimensional arrays for Python.

barecat. Efficient data storage format optimized for random-access reads, especially in machine learning workflows

mlflow. The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while controlling costs and managing access to models and data.

datamodel-code-generator. Generate Pydantic v2 models, dataclasses, TypedDict, and msgspec.Struct from OpenAPI, JSON Schema, GraphQL, Avro, Protobuf, and raw JSON/YAML/CSV.

DiffusionSfM. [CVPR 2025] "DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion" official implementation.

onedrive-downloader. Shell

tokei. Count your code, quickly.

rooster. Release management tool

ty. An extremely fast Python type checker and language server, written in Rust.

cinny. Yet another matrix client

pyrefly. A fast type checker and language server for Python

styleguide. Style guides for Google-originated open-source projects

Trainer. 🐸 - A general purpose model trainer, as flexible as it gets

capi. Code and weights for the paper "Cluster and Predict Latents Patches for Improved Masked Image Modeling"

sps. Rust based package manager for macOS

llm-mlx. Support for MLX models in LLM

perception_models. State-of-the-art Image & Video CLIP, Multimodal Large Language Models, and More!

MatrixCity. [ICCV 2023] MatrixCity: A Large-scale City Dataset for City-scale Neural Rendering and Beyond.

codex. Lightweight coding agent that runs in your terminal

void. TypeScript

ruff-api. Experimental Python API for Ruff

pytorch-accelerated. A lightweight library designed to accelerate the process of training PyTorch models by providing a minimal, but extensible training loop which is flexible enough to handle the majority of use cases, and capable of utilizing different hardware options with no code changes required. Docs: https://pytorch-accelerated.readthedocs.io/en/latest/

diffusers. 🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.

GeometryCrafter. [ICCV 2025] GeometryCrafter: Consistent Geometry Estimation for Open-world Videos with Diffusion Priors

data. A PyTorch repo for data loading and utilities to be shared by the PyTorch domain libraries.

MegaLoc. An image retrieval model for any localization task

gradio. Build and share delightful machine learning apps, all in Python. 🌟 Star to support our work!

anycam. Official repository for "AnyCam: Learning to Recover Camera Poses and Intrinsics from Casual Videos" (CVPR 2025)