Audit Your GPU Infrastructure with AI Agents

Ask your AI agent for a full inventory of your GPU cloud. It lists every resource, calculates costs, identifies waste, and suggests optimizations.

---

​

”Give me a complete audit of everything running in my workspace.”

The agent inventories all resources:

1. list_instances — all GPU instances
2. list_volumes — all storage volumes
3. list_environments — all environments
4. billing_balance — current credit balance
5. billing_usage — spending over the last 30 days

Infrastructure Audit — May 19, 2026 Instances (4 running):

Instance	Environment	GPU	Uptime	Cost/hr
prod-inference	production	H100	12 days	$2.50
staging-test	staging	A100	3 days	$1.60
dev-box	main	RTX 4090	18 hrs	$0.35
old-benchmark	main	RTX 4090	9 days	$0.35

Volumes (3 total):

Volume	Environment	Size	Region	Status
model-weights	production	500 GB	us-east-1	attached
training-data	main	200 GB	us-east-1	available
temp-scratch	main	100 GB	eu-central-1	available

Environments: main, staging, production Balance: $1,247.80 **30-day spend:**$1,892.40 Current burn rate: $4.80/hr ($115.20/day)

---

​

“What’s wasting money? Where can I cut costs?”

The agent analyzes utilization:

1. ssh_execute on each instance — checks GPU utilization
2. list_volumes — checks for unattached volumes

Waste identified:

Resource	Issue	Monthly cost	Action
old-benchmark	0% GPU utilization for 9 days	$75.60	Terminate
staging-test	2% GPU utilization	$115.20/mo	Terminate when not testing
temp-scratch	Unattached volume, different region	$7.50/mo	Delete if not needed
training-data	Unattached volume	$15.00/mo	Delete or attach

Potential savings: $213.30/month (45% of current spend) Kill old-benchmark first (idle 9 days), then shut down staging-test between sessions, and delete temp-scratch (wrong region).

---

​

”Clean up the waste. Kill idle instances and delete unused volumes.”

The agent executes the cleanup:

1. delete_instance — terminates old-benchmark
2. delete_volume — deletes temp-scratch
3. billing_balance — checks updated balance after refunds

Cleanup complete:

• Terminated old-benchmark (idle 9 days) — saving $0.35/hr
• Deleted temp-scratch (100 GB, eu-central-1) — unused credits refunded

Updated burn rate: $4.45/hr (was$4.80/hr) Monthly savings: ~$252 Remaining resources: 3 instances, 2 volumes across 3 environments. Next recommendation: shut down staging-test when you’re not actively testing.

---

​

Tools used in this workflow