Company
Built by people who used to wait for GPUs
We started Ferrite in 2022 after one too many training runs stuck in a queue. Today 120 of us run 14,000 GPUs across seven regions, and nobody waits three weeks for a cluster.
How we work
Three things we don’t compromise on
Uptime is the product
A fast cluster that falls over is a slow cluster. We’d rather ship one feature that stays up than three that don’t.
Engineers answer support
Every support ticket is read by someone who can SSH into the machine. No scripts, no tiers.
Say the price
Every price is on the website. If you need a calculator to understand your bill, we’ve failed.
Team
The people on call
AL
Ava Lindqvist
Co-founder, CEO
TR
Tomás Reyes
Co-founder, CTO
Kenji Mori
Advisor
PN
Priya Natarajan
VP Infrastructure
SO
Sam Okafor
Head of Networking
LB
Lena Brandt
Head of Support
MH
Marcus Hale
Data Hall Lead, ams-1
YP
Yuna Park
Head of Research
Careers
Open roles
Competitive salary and equity, 30 days off, and a hardware budget you’ll actually use.
Senior Network Engineer
Infrastructure
Data Hall Technician
Infrastructure
Distributed Systems Engineer
Engineering
Scheduler Engineer (Rust)
Engineering
Frontend Engineer, Console
Engineering
ML Solutions Engineer
Customers
Support Engineer
Customers
Research Engineer, Accelerators
Research
Start in minutes
Your first 100 GPU-hours are on us
Sign up, add a card and launch a cluster before your coffee’s cold. No sales call, no minimum spend.