aashanm

aashanm / arboric

Public

Rescheduling AI workloads to save cost and energy

arboric.xyz carbon-aware cli cloud-infrastructure finops gpu-orchestration
28
6
100% credibility
Found Feb 17, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Arboric helps schedule energy-intensive workloads like AI training to times of low electricity cost and carbon emissions using simulated grid forecasts.

How It Works

1
🌲 Discover Arboric

You learn about Arboric, a friendly tool that smartly times your heavy computer tasks to run when energy is cheap and clean.

2
📦 Set it up

With one easy command, you bring Arboric to your computer, and it's ready to help right away.

3
⚙️ Pick your balance

You choose how much to prioritize saving money or reducing pollution, just like setting a simple dial.

4
🚀 Run a demo

You try the demo and see sample tasks magically shifted to perfect times, with colorful results popping up.

5
🔍 Plan your task

You describe your job – its length and power needs – and Arboric finds the best window within your deadline.

6
📊 View savings

Beautiful tables and charts show exactly how much money and pollution you'll save compared to running now.

💚 Go greener

Your tasks now run during clean energy times, cutting your bills and carbon footprint effortlessly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 28 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is arboric?

Arboric reschedules Python-based AI workloads—like LLM training or ETL pipelines—to low-cost, low-carbon grid windows, slashing electricity bills and emissions without code changes. You define a job's duration, power draw, and deadline via CLI (`arboric optimize "Job" --duration 6 --power 120`), and it scans forecasts for US-WEST, EU-WEST, or other regions to suggest optimal start times. A companion FastAPI server exposes REST endpoints for Airflow or Prefect integration, plus JSON/CSV exports for analysis.

Why is it gaining traction?

Its polished Rich-powered CLI delivers instant tables comparing baseline vs. optimized runs (e.g., 41% cost savings, 59% CO2 cut), making green scheduling feel effortless amid rising cloud energy costs. Zero-config defaults and config YAML for weights (70% cost/30% carbon) hook devs fast, while fleet demos simulate real AI queues. Multi-region duck curve modeling and event detection (price spikes, green windows) add practical edge over basic cron tools.

Who should use this?

ML engineers batching GPU training jobs, data teams delaying ETL or analytics pipelines, and DevOps handling serverless workloads across cloud-to-edge setups. Ideal for anyone in arboriculture & urban forestry firms optimizing compute or Python scripters chasing energy savings on flexible batch runs.

Verdict

Try the CLI demo now—docs and PyPI install are solid for alpha (16 stars, 1.0% credibility)—but hold for production until WattTime integration lands and tests expand. Great for cost/energy-aware prototyping, less for mission-critical fleets.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.