Senior Inference Reliability Engineer

Parasail•San Mateo, CA

11d

About The Position

Parasail is redefining AI infrastructure by enabling seamless deployment across a distributed network of GPUs, optimizing for cost, performance, and flexibility. Our mission is to empower AI developers with a fast, cost-efficient, and scalable cloud experience—free from vendor lock-in and designed for the next generation of AI workloads. The Senior/Staff Inference Reliability Engineer will own the end-to-end reliability and production performance of customer inference workloads. This role sits at the intersection of inference platform engineering, LLM performance, and infrastructure reliability. You will ensure that customer endpoints meet expectations for availability, latency, throughput, quality, and cost. When an endpoint degrades, you will follow the problem across the entire serving path—from APIs, routing, scheduling, and autoscaling through model servers, GPUs, networking, and underlying infrastructure—and drive it through resolution. This is not a traditional DevOps role focused only on clusters and deployments. It is a production systems role for an engineer who enjoys investigating ambiguous performance problems, building diagnostic tooling, and turning recurring incidents into durable platform improvements. Prior LLM-inference experience is valuable but not required. We are looking for someone with deep production systems experience who can quickly learn inference-specific technologies and metrics.

Requirements

5+ years of experience in production engineering, site reliability engineering, infrastructure engineering, distributed systems, ML infrastructure, database reliability, or performance engineering.
Demonstrated ownership of a critical production service or workload.
Experience diagnosing complex latency, throughput, capacity, or reliability problems across multiple system layers.
Strong software-engineering ability beyond infrastructure configuration and CI/CD automation.
Hands-on experience building observability, automation, diagnostic tooling, or production safeguards.
Strong communication and technical leadership skills, including the ability to coordinate incident resolution across engineering teams.
Experience with Kubernetes, Linux, networking, and cloud-native infrastructure.
A demonstrated ability to learn unfamiliar systems and develop deep technical expertise is essential.

Nice To Haves

Prior LLM-inference experience is valuable but not required.
Experience with GPUs, ML infrastructure, model serving, vLLM, SGLang, Triton, TensorRT-LLM, or similar technologies is valuable but not required.
Experience operating across multiple cloud providers, regions, hardware configurations, or infrastructure suppliers is especially valuable.

Responsibilities

Own the production health of customer inference workloads, including availability, request success, time to first token, inter-token latency, throughput, and operational efficiency.
Establish clear service-level indicators, objectives, performance baselines, and escalation paths for production endpoints.
Build the telemetry, dashboards, alerts, and automated diagnostics needed to detect meaningful endpoint degradation before customers report it.
Create visibility across the full inference-serving path, including request queues, routing, scheduling, model servers, GPU utilization, networking, storage, and provider infrastructure.
Lead the investigation of complex latency, throughput, capacity, and reliability regressions.
Determine whether an issue originates in customer traffic patterns, platform services, inference-engine configuration, GPU hardware, networking, storage, or an external infrastructure provider.
Remain accountable for the customer outcome while partnering with the appropriate engineering teams to implement the fix.
Help lead customer-impacting incidents and establish effective operational practices for acknowledgement, diagnosis, recovery, and communication.
Convert significant incidents into automated tests, safeguards, runbooks, capacity controls, anomaly detection, and platform improvements.
Partner with the LLM Performance team to validate that engine-level optimizations deliver measurable improvements in production.
Analyze workload behavior, capacity requirements, utilization, tail latency, and cost efficiency across heterogeneous GPU providers and hardware.
Help ensure that customer performance requirements are met without consuming unnecessary infrastructure capacity.
Identify recurring patterns across incidents, workloads, and customer escalations.
Translate those findings into improvements to the inference platform, reliability architecture, deployment processes, observability, and product roadmap.