Overview As Principal Engineer for the Inference Cloud Platform, you translate ambitious scale and reliability goals into a long-term architecture. You will own multi-region topology, failure domains, and system evolution while delivering fast, available inference services. You'll lead across ML, Product, and Infrastructure teams to solve high-leverage platform problems and drive execution when specs are unclear. Expect to shape platform decisions, production code on critical paths, and strong operational rigor in a rapidly growing setting.
Responsibilities- Define and prioritize core platform problems and trade-offs for the Inference Cloud Platform
- Set long-term technical direction, topology, and system boundaries for multi-region services
- Architect reliable, low-latency systems with active-active failover and graceful degradation
- Contribute production code on critical paths and review designs with long-term operational impact
- Lead production issues, drive observability, incident response, and capacity planning
- Influence cross-functional teams on API design, deployment strategy, and shared infrastructure decisions
- Mentor teams to improve technical decision-making and engineering standards
Key requirements- 10+ years of software engineering experience
- Strong distributed systems and cloud infrastructure background
- Proven ability to design and operate highly available, latency-sensitive systems at scale
- Experience optimizing latency, throughput, and efficiency in high-QPS workloads
- Production-ready coding ability in Go, C++, or Python
- Experience with observability practices (metrics/logging/tracing) and SLO/SLA-driven operations
- Ability to influence senior engineers and cross-functional partners through credibility and judgment
- Experience with ML inference infrastructure, model serving, or GPU-accelerated workloads is a plus
- strong technical judgment
- excellent communication
- leadership and mentorship
- distributed systems architecture in cloud environments
- networking, compute orchestration, container platforms
- multi-region production services
