Senior MLOps Engineer – Artificial Intelligence (up to $290k)

Dex; New York (NY); On-site; USD 150,000 - 185,000; Full time

Job summary

Dex partner team seeks an experienced MLOps-focused Python engineer to design and build scalable ML infrastructure for continuous training, inference, and monitoring at scale.

You will collaborate with AI Platform and ML engineers, delivering robust systems with high availability for production use across critical, high-volume environments.

Qualifications

• 5+ years of Python in production ML environments.
• Experience operating or building ML infrastructure with Kubernetes and Argo Workflows.
• Knowledge of ML frameworks such as PyTorch, ONNX, or DeepSpeed; scale considerations.
• Experience with at least one major cloud provider (AWS, GCP, or Azure).
• Strong CS fundamentals and track record of production-quality code.

Responsibilities

• Design and build continuous training pipelines for rapid iteration and deployment of ML models.
• Architect and implement inference infrastructure for high throughput and low latency products.
• Develop monitoring workflows and define SLAs for latency, throughput, and resource usage.

Skills

Python

Production ML

CS fundamentals

System design

Cloud concepts

Tools

Kubernetes

Argo Workflows

PyTorch

ONNX

DeepSpeed

Job description

This role is with one of Dex’s trusted partner companies. We work closely with their teams to truly understand their culture, goals, and what they’re looking for, so we can match you with the right opportunity and give you context about the role before you commit to a process.

The role

We’re talking about a team with a long history of shipping production AI, processing vast and complex financial datasets that underpin global capital markets. They build search, discovery, and workflow products on top of advanced models, serving hundreds of thousands of users who depend on real-time, reliable systems.

You’ll join a specialized MLOps team, owning the tooling and infrastructure that keeps this model development lifecycle reliable, fast, and observable. This role is about designing and building the core systems for continuous training, inference, and monitoring at a scale few companies can offer. It’s not about ad-hoc scripting or managing a handful of models; it’s about architecting robust, high-SLA platforms for a critical, high-volume environment.

The work

• Design and build continuous training pipelines that enable rapid iteration and confident deployment of ML models.
• Architect and implement inference infrastructure capable of handling high throughput and low latency demands for critical user-facing products.
• Develop comprehensive monitoring workflows and define strict SLAs around latency, throughput, and resource usage (CPU, GPU, memory, network).
• Collaborate with AI Platform teams to operationalize models end-to-end, ensuring seamless integration from research to production.
• Partner directly with ML engineers building customer-facing products to understand their needs and deliver robust, scalable MLOps solutions.

What You Bring

• 5+ years of professional experience as a strong Python developer, with hands‑on work in production ML environments.
• Experience operating or building ML infrastructure using cloud-native tooling like Kubernetes and Argo Workflows.
• Working knowledge of ML frameworks such as PyTorch, ONNX, or DeepSpeed, and an understanding of operational demands at scale.
• Experience with at least one major cloud provider (AWS, GCP, or Azure) and strong reasoning about infrastructure decisions.
• Solid CS fundamentals (data structures, algorithms, system design) and a track record of delivering production-quality code.