ML Inference Platform Engineer — Low-Latency GPU Systems

Dialpad; Buenos Aires; Híbrido; ARS 133.982.000 - 193.530.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Competitive salary

Comprehensive benefits

Opportunities for growth

AI tools training

Descripción de la vacante

Dialpad is hiring a Software Engineer for the ML Inference Platform to build production systems that serve in-house AI models at scale. You will work at the intersection of model development, runtime systems, and cloud infrastructure on NVIDIA GPUs in GCP.

This is an implementation-heavy role focused on model serving, deployment safety, telemetry, and cost-aware optimization. You’ll help define packaging, deployment, benchmarking, and operations across environments.

Formación

• 6+ years of professional software engineering experience shipping backend services or production platforms.

Responsabilidades

• Design, build, and improve systems connecting AI capability development to production inference.
• Inference Serving & Runtime Systems for low-latency, high-throughput workloads.
• GPU Infrastructure & Utilization on Kubernetes/GCP with NVIDIA GPUs.
• Model Server Integration with frameworks like vLLM, Triton, or TGI.
• Traffic, release safety, and canary rollout mechanisms for model-backed services.
• Benchmarking & evaluation tooling for latency, throughput, cost, and reliability.
• Artifact lifecycle management across environments and observability capabilities.

Conocimientos

Production engineering

Python/Go coding

Distributed systems

Kubernetes & Linux

Observability & debugging

Collaborative communication

Herramientas

Kubernetes

Linux

CI/CD

Triton

vLLM

GCP

Descripción del empleo

Dialpad is hiring a Software Engineer for the ML Inference Platform to build production systems that serve in-house AI models at scale. You will work at the intersection of model development, runtime systems, and cloud infrastructure on NVIDIA GPUs in GCP.

This is an implementation-heavy role focused on model serving, deployment safety, telemetry, and cost-aware optimization. You’ll help define packaging, deployment, benchmarking, and operations across environments.