# Xiangyue-Zhang/auto-deep-researcher-24x7

🔥 An autonomous AI agent that runs your deep learning experiments 24/7 while you sleep. Zero-cost monitoring, Leader-Worker architecture, constant-size memory.

Repository: https://github.com/Xiangyue-Zhang/auto-deep-researcher-24x7
Canonical: https://ross.abutalabs.com/products/auto-deep-researcher-24x7
Homepage: https://arxiv.org/abs/2604.05854
Language: Python
License: Apache-2.0
License Family: permissive
Topics: ai-agent, autonomous-agent, claude-code, deep-learning, experiment-automation, gpu, hyperparameter-tuning, llm-agent, machine-learning, mlops, pytorch, research-automation
Last push: 2026-06-03T03:18:50+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 85, release rhythm 35, longevity 10
- inputs: {"age_days": 147, "days_push": 91, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1283, forks 112 (observed 2026-08-28T04:04:14.388862+00:00)

## What it is
An open-source Python framework where LLM agents autonomously run deep learning experiments 24/7, covering hypothesis formation, code implementation, training, and analysis. It uses a Leader-Worker multi-agent architecture, constant-size memory, and zero-cost process-level monitoring to keep long-running autonomous experimentation cheap.

## Use cases
- run deep learning experiments autonomously overnight
- automate hyperparameter tuning with an LLM agent
- monitor GPU training jobs without paying for LLM API calls
- run ML experiments on a Slurm cluster automatically
- iterate on research experiments 24/7 while I sleep
- keep agent memory bounded during long autonomous runs

## When to choose
- you want fully autonomous end-to-end deep learning experiment cycles
- you need low-cost long-running agent deployments (days to weeks)
- you train PyTorch models locally or on Slurm clusters
- you want to plug in Claude Code, Codex CLI, or OpenAI-compatible LLM APIs like DeepSeek or Qwen

## When to avoid
- you need a general-purpose coding assistant rather than an experiment loop
- your workloads are not deep learning training runs
- you require a managed SaaS with a web dashboard rather than a local Python framework
- you cannot grant an agent permission to execute training code on your machine or cluster

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, machine-learning, monitoring, workflow-automation, scheduling
- domain: artificial-intelligence, machine-learning, deep-learning, developer-tools
- platform: python, cli
- tags: autonomous-agent, experiment-automation, mlops, hyperparameter-tuning, pytorch, claude-code, leader-worker-architecture, zero-cost-monitoring, slurm, research-automation, ai-agents, automation, linux, gpu

## Member repositories
- Xiangyue-Zhang/auto-deep-researcher-24x7 (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:14.388862+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:56:47.535964+00:00, confidence not recorded.
  - readme: https://github.com/Xiangyue-Zhang/auto-deep-researcher-24x7 (fetched 2026-08-28T04:04:14.388862+00:00, sha 25b991c2739c)
  - homepage: https://arxiv.org/abs/2604.05854 (fetched 2026-08-29T12:12:45.031957+00:00, sha 50cc067a6af4)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T12:12:45.041398+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T12:12:45.044884+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T12:12:45.046800+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T12:12:45.043219+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.