Production ML Engineer: Kubernetes, Triton, Python

Soilair Selection; Fort Meade (MD); On-site; USD 150,000 - 190,000; Full time

Benefits offered by this job

Medical coverage

Dental coverage

Vision coverage

Health Savings Accounts

Flexible Spending Accounts

Flexible work schedule

401(k)

Company-funded life insurance

Paid Time Off (PTO)

Tuition and training assistance

Employee referral rewards

Pet insurance

Employee discount program

Job summary

Soilair Selection seeks a skilled software engineer to move VISTA models from research prototypes to production inference services using Kubernetes and NVIDIA Triton. You will build CI/CD pipelines, tune for throughput, and develop thorough user documentation.

Responsibilities include ensuring scalable deployment pipelines, optimizing performance, and delivering clear documentation for end users and operators in a TS-cleared environment.

Qualifications

• Requires a Top Secret clearance with Full Scope poly.
• Bachelor's degree in Computer Science or 14+ years of software engineering experience.
• Strong Python skills.
• Experience with containerization and orchestration tools.

Responsibilities

• Move VISTA models from research prototypes to production inference services using Kubernetes and NVIDIA Triton.
• Build CI/CD pipelines and tune for throughput.
• Create comprehensive user documentation.

Skills

Python

DevOps

CI/CD

Docker

Kubernetes

NVIDIA Triton Inference Server

TensorRT

C++

Java

GoLang

CUDA

Natural Language Processing

Computer Vision

Automated Speech Recognition

ONNX

Infrastructure-as-code

Education

Bachelor's degree in Computer Science

Tools

Docker

Kubernetes

NVIDIA Triton Inference Server

TensorRT

ONNX

Job description

Soilair Selection seeks a skilled software engineer to move VISTA models from research prototypes to production inference services using Kubernetes and NVIDIA Triton. You will build CI/CD pipelines, tune for throughput, and develop thorough user documentation.

Responsibilities include ensuring scalable deployment pipelines, optimizing performance, and delivering clear documentation for end users and operators in a TS-cleared environment.