ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

Run a Parallel Autonomous ML Research Organization on your OpenClaw instance.

Research Agora: Claude Code skills, benchmarks & tools for ML researchers — paper writing, citation verification, experiment tracking, LaTeX automation

Code for the paper "2Mamba2Furious: Linear in complexity, competitive in accuracy"

Automated Framework for Evaluating Chain-of-Thought Reasoning in LLMs.

Enhance research paper writing by organizing workflows, improving structure, and supporting self-review for clear, logical, and well-aligned content.

Hierarchical GNN for fake-news detection on NELA-GT-2022: multi-level graphs (word, sentence, article + metadata) + MLflow, baselines & explainability.

Claude Code skill for autonomous ML research — from topic to LaTeX paper

Benchmarking PEFT (LoRA vs adapters) for ECG segment classification using Wav2Vec2, HuBERT, and ECG-FM with full metric reporting and ablations.

Hypothesis-first experiment tracking for agent-driven ML research, wired into Claude Code.

Autonomous ML research agent skill (Agent Skills format). Designs experiments, deploys to Modal GPUs, tracks budget, iterates autonomously.

This repository presents four Capstone projects developed as part of the "the AI Engineer" program

This is a comprehensive analysis of 5 HPO algorithms- General Algorithms (GA), Particle Swarm Optimization (PSO), (DE), PyHopper HPO, Bayesian Optimization and HyperBand Optimization (BOHB),

Causal analysis framework using Double Machine Learning to quantitatively isolate the effect of model size on deep learning performance while controlling for confounders such as dataset size, training time, and hyperparameters.

A simplified three-phase research workflow for Claude Code: Discovery, Experiment, Paper

Provider-neutral autonomous ML research paper writer — ideation -> LaTeX writeup -> peer review -> anti-hallucination numerical audit. Spun out from hermes-sci to remove Hermes-runtime coupling.

Parametric Sigmoid Linear Unit

Julia implementation of ST-ProtoPNet for interpretable classification with support and trivial prototypes. Built on Flux.jl with custom losses, data prep tools, and 2D visualization.

MemorySafe - 99.4% catastrophic forgetting reduction

A curated collection of key research papers and resources on LLMs and deep learning, organized by topic to provide a clear learning path from fundamentals to advanced techniques.