Remote Research Staff: Voice AI Foundations

Deepgram; Myrtle Point (OR); Remote; USD 150,000 - 210,000; Full time

Job summary

Deepgram is seeking a Research Staff member to pioneer Latent Space Models for scalable, cost-efficient voice AI. You will tackle neural audio codecs, steerable generative models, and embedding systems enabling large-scale data synthesis and multimodal speech-to-speech systems.

You will work remotely with a focus on bridging theory and practical deployment, delivering robust research with real-world impact and collaborating across AI foundations teams.

Qualifications

• Strong mathematical foundation in statistical learning theory and related areas.
• Deep expertise in foundation model architectures and scalable training across modalities.
• Proven ability to bridge theory and practice and implement novel formulations efficiently.
• Demonstrated ability to build data pipelines for large-scale, diverse datasets.
• History of open-source contributions or research publications advancing voice/AI.
• Experience designing controlled experiments with rigorous ablations and metrics.

Responsibilities

• Pioneer development of Latent Space Models to address data, scale, and cost for voice AI.
• Research and design of neural audio codecs and steerable generative models.
• Develop embedding systems to factorize latent representations for control and data augmentation.
• Explore synthetic data generation and multimodal speech-to-speech systems at scale.
• Design architectures and training schemes optimized for hardware and real-time inference.

Skills

Math foundations

Foundation models

Experiment design

Data pipelines

Open-source contributions

Job description

Deepgram is seeking a Research Staff member to pioneer Latent Space Models for scalable, cost-efficient voice AI. You will tackle neural audio codecs, steerable generative models, and embedding systems enabling large-scale data synthesis and multimodal speech-to-speech systems.

You will work remotely with a focus on bridging theory and practical deployment, delivering robust research with real-world impact and collaborating across AI foundations teams.