Tools & Resources
Tools I have built and practical guides I share for researchers working with AI and high-performance computing. Everything is free and open source on GitHub.
Tools I Built
Write the notes, the equations and the to-do list together, live.
A shared notebook for research groups. Open a link and everyone types into the same page at once, with LaTeX math that renders as you write. It runs on your own server, so your notes stay with you.
- Real-time collaborative editing with live cursors
- Markdown + LaTeX with an instant rendered preview
- Self-hosted: your notes never leave your server
ImgMorph
A lightweight desktop app that converts images (HEIC, PNG, BMP, TIFF, WebP and more) to JPG, with drag-and-drop, batch conversion and a progress bar.
Cheat Sheets & Tutorials
A quick reference for SLURM: writing sbatch scripts, checking running and pending jobs, partitions and GPU usage, priority and fairshare, and job history.
A hands-on guide to training deep learning models on GPU clusters with SLURM, from HPC basics to single-GPU, multi-GPU (DDP) and multi-node training, federated learning at scale, and monitoring and debugging.
Training scripts for federated learning (FedAvg, CIFAR-10 / ResNet-18) on the UPPMAX cluster, from a single-GPU baseline to memory-efficient multi-GPU runs, packaged with Apptainer.