Logistics · data sync

One customer’s nightly import used to block everyone’s morning syncs. Fair queueing per account fixed it.

p95 sync wait 38 min → 2 min

900 customer accounts

1 line of config

A concurrency key per account fixed it in one line.

Priya Raman

Senior SRE

Quanta Freight

Quanta Freight case study cover

Where they started

Quanta syncs shipment data for 900 customer accounts. One large customer imported 400,000 rows every night, and smaller customers’ morning syncs queued behind it.

What they built

Syncs moved to a workflow with a concurrency limit keyed on the account ID. The runtime serves accounts in turn, so the big import still finishes overnight while small syncs start within seconds.

Long imports now sleep between batches of 5,000 rows instead of holding a worker, which cut their compute bill as a side effect.

What changed

The 95th-percentile wait for a sync fell from 38 minutes to under two, and the fix was one line of configuration.

Get started

Ship the agent. Keep the receipts.

Free for 50,000 steps a month. No credit card, no separate workers, and your first durable workflow deployed before lunch.

orrindel

Durable runtime for AI agents and automations. Every run, in plain text.

Book a 20-minute demo →
All systems normal
99.99% uptime · last 90 days
90 days agotoday
regions us-east · eu-west · ap-south
soc 2 type ii · gdpr · hipaa (baa)
sdk v4.2.1 · node · python · go
© 2026 Orrindel Labs, Inc.PrivacyTermsSecurityMade in plain text.

Create a free website with Framer, the website builder loved by startups, designers and agencies.