Senior ML Platform Engineer - Remote, High-Impact Systems

Bright Vision Technologies; Woodbridge Township (NJ); Remoto; USD 100.000 - 160.000; Tempo pieno

Descrizione del lavoro

Bright Vision Technologies is seeking an ML Platform Engineer to design, build, and operate high-performance inference platforms for production ML models. This 100% remote role emphasizes distributed systems, latency vs.

throughput trade-offs, and end-to-end reliability at scale. You will apply Python and a systems language (Go/Rust/C++) with Kubernetes and modern cloud platforms to deliver scalable, cost-efficient serving.

Competenze

• 10+ years of experience in distributed systems, infra, or ML platform engineering.
• Strong proficiency in Python and a systems language (Go/Rust/C++).
• Production-grade high-throughput, low-latency services.
• Experience with LLM inference frameworks such as vcLLM or TensorRT-LLM.
• Kubernetes and modern cloud platforms.
• Observability stacks (metrics, tracing, structured logging).
• Strong communication and incident response skills.

Mansioni

• Design and operate model serving platforms supporting diverse workloads.
• Optimize inference performance with batching, caching, and routing.
• Implement multi-tenant routing, rate limiting, and QoS policies.
• Build autoscaling and capacity management systems balancing latency and cost.
• Tune GPU utilization and memory for LLM serving workloads.
• Integrate serving with API gateways and observability platforms.
• Develop deployment workflows with canaries and automated rollbacks.
• Drive incident response and durable reliability improvements.
• Collaborate with ML and product teams on model releases.

Conoscenze

Distributed systems

Python

Go

Rust

C++

Kubernetes

Cloud platforms

Observability

LLM frameworks

Formazione

CS degree

Strumenti

vcLLM

TensorRT-LLM

Descrizione del lavoro

Bright Vision Technologies is seeking an ML Platform Engineer to design, build, and operate high-performance inference platforms for production ML models. This 100% remote role emphasizes distributed systems, latency vs.

throughput trade-offs, and end-to-end reliability at scale. You will apply Python and a systems language (Go/Rust/C++) with Kubernetes and modern cloud platforms to deliver scalable, cost-efficient serving.