About the role
As generative media reshapes industries across a market projected to grow by hundreds of billions over the next decade, fal is becoming the ecosystem that ambitious teams build on.
About this role
You are an experienced software engineer who is passionate about building large-scale computing platforms. You have experience building systems that remain reliable under high traffic, partial failures, and changing capacity. You know how to deliver reliability, performance, and scale with minimum operational load.
What you'll do
Take ownership of one or more of the Rust/Python systems that power our AI inference and large-scale GPU computing platform
Design for 100x growth in workloads and GPU capacity, with reliable, low-latency execution across the globe
Use AI aggressively to accelerate development and automate repetitive operations, alerting, and recovery
You will specialize in one or more of these critical areas:
Request routing and queuing on a global scale
App deployment and autoscaling
Worker orchestration and placement
GPU fleet autoscaling and capacity management
GPU fleet lifecycle automation: provisioning, diagnostics & recovery, upgrades, tracking
Global file storage, caching and distribution
Global IP backbone: private networking across datacenters, elastic public IPs, ACLs/firewalls, private connectivity
Workload execution: containerization, globally distributed filesystems, RAM/VRAM snapshotting
Qualifications/Nice to have
3+ years building and operating large-scale production systems, with a track record of reliability and scale
Strong Rust and/or Python skills
Strong systems fundamentals, with technical depth relevant to your specialization: distributed coordination, scheduling, fault tolerance, Linux, networking, storage, or capacity management
Experience building and using observability to drive performance and reliability decisions
Clear communication, sound technical judgment, and the initiative to move quickly, drive decisions across teams, and own systems from design through production
Multi-tenant compute platforms, AI inference or training infrastructure, GPU workload scheduling
High-performance systems programming: async runtimes, zero-copy, memory-safe concurrency
Linux systems engineering and fleet automation: configuration management, kernel and driver debugging, safe upgrades, and automated recovery
Distributed filesystems like JuiceFS and Lustre
Experience with Nvidia/AMD GPU infrastructure; DCGM, NVLink, RDMA, InfiniBand/RoCEv2
Global networking and routing: BGP, ECMP, tunnels, eBPF, Geneve