ML Inference Platform Engineer: Low-Latency AI on GPUs

Dialpad; Buenos Aires; Presencial; ARS 181.094.000 - 271.641.000; Jornada completa

Descripción de la vacante

Dialpad is seeking ML Inference Platform Engineers to build the production systems that serve our in-house AI models at scale. This role lies at the intersection of model development, high-performance runtime systems, and cloud infrastructure.

You will help turn trained models and AI capabilities into reliable, observability-rich, low-latency services running on NVIDIA GPUs in GCP. This is an implementation-heavy role focused on inference machinery, deployment safety, and scalable production

Formación

• 6+ years of professional software engineering experience shipping backend services, infrastructure systems, or production platforms.
• Proficiency in writing maintainable production code in Python, Go, or another backend-oriented language.
• Experience building, operating, or optimizing high-throughput services, distributed systems, or ML infrastructure.
• Hands-on experience with containers, Kubernetes, Linux environments, CI/CD, deployment automation, and production operations.
• Strong instinct for reproducibility, observability, rollout safety, failure modes, and resilience.
• Comfort reasoning about bottlenecks across compute, memory, network, storage, batching, concurrency, and SLOs.
• Ability to work closely with model developers, product engineers, infrastructure teams, and leadership.

Responsabilidades

• Design, build, and improve systems connecting AI capability development to production inference.
• Develop inference serving and runtime systems for low-latency, high-throughput workloads.
• Operate and optimize GPU-backed containerized workloads on Kubernetes/GCP.
• Integrate model-serving frameworks and runtimes (e.g., vLLM, Triton, TGI) for internal deployment and observability.
• Enable shadow serving, canary rollouts, and safe rollback mechanisms for model-backed services.
• Build tooling to measure latency, throughput, cost, and reliability under production traffic.
• Improve packaging, versioning, deployment, and rollback of model artifacts across environments.
• Strengthen telemetry, logging, tracing, dashboards, and alerting for production insight.
• Contribute to compute efficiency and cost-performance tradeoffs across the platform.

Conocimientos

Production Engineering Experience

Strong Software Fundamentals

Inference or Systems Orientation

Kubernetes & Linux Fluency

Operational Judgment

Performance Awareness

Collaboration

Herramientas

vLLM

Triton

TGI

Descripción del empleo

Dialpad is seeking ML Inference Platform Engineers to build the production systems that serve our in-house AI models at scale. This role lies at the intersection of model development, high-performance runtime systems, and cloud infrastructure.

You will help turn trained models and AI capabilities into reliable, observability-rich, low-latency services running on NVIDIA GPUs in GCP. This is an implementation-heavy role focused on inference machinery, deployment safety, and scalable production