Applied ML Engineer: From Research to Production

deepgram; United States; On-site; USD 140,000 - 180,000; Full time

Job summary

Deepgram is seeking an Applied ML Engineer to own and streamline the research-to-production pipeline for real-time speech models. You will work shoulder-to-shoulder with researchers to move models from research notebooks to production systems, focusing on robust training, evaluation, packaging, and safe deployment.

The role spans our hybrid infrastructure, covering GPU data centers and cloud, and relies on in-house tooling that makes translating ideas into production models fast and reliable.

Responsibilities

• Own and streamline the research-to-production pipeline for models.
• Collaborate with research scientists to take models from checkpoint to production safely.
• Hardening training and evaluation workflows, building packaging and deployment paths for production use.
• Ensure production models run across a hybrid training and inference stack (in-house GPUs, data centers, and cloud).
• Leverage in-house tooling to turn research ideas into production models without rewriting code.

Job description

Deepgram is seeking an Applied ML Engineer to own and streamline the research-to-production pipeline for real-time speech models. You will work shoulder-to-shoulder with researchers to move models from research notebooks to production systems, focusing on robust training, evaluation, packaging, and safe deployment.

The role spans our hybrid infrastructure, covering GPU data centers and cloud, and relies on in-house tooling that makes translating ideas into production models fast and reliable.