Backend Infrastructure Engineer - K3s & Bare Metal
15M users. Billions of messages. 100k IOPS on Postgres. Hundreds of GPUs on K3s.
Growing fast and needing to scale smart.
The Challenge: Building infrastructure for the next 100M users. We've scaled from 0 to 15M in
under 2 years - now it's time to architect for the next phase.
What You'll Build:
●
Next-gen GPU orchestration on K3s for hundreds of GPUs
S3 archival pipeline to optimize our 100k IOPS Postgres setup
Multi-region K3s clusters on bare metal with Tailscale mesh
Advanced queue systems with intelligent batching
Cloudflare tunnels for global traffic optimization
Monitoring that predicts issues before they happen
Core Projects:
●
Architect multi-region infrastructure for global scale
Optimize GPU scheduling for AI workloads at scale
Build data lifecycle management (hot storage → S3)
Design queue systems that handle billions of messages
Create infrastructure tooling for 10x developer velocity
Experience:
●
3+ years scaling production infrastructure
Deep K3s/K8s expertise
PostgreSQL optimization for high IOPS environments
Distributed systems architecture experience
Bare metal + Linux networking mastery
Track record of building reliable systems at scale
Why Janitor AI: Started in 2023, viral growth to 15M users. Building the entertainment/creativity
platform that ChatGPT is for productivity. Real scale, real challenges, real impact on millions of
daily users.