Senior ML Research Scientist (Speech)

United States

Apply with AI Apply

$170,000 – $220,000Compensation

Senior (5 to 8 years)Experience Level

Full TimeJob Type

UnknownVisa

Healthcare, Artificial Intelligence, Medical ImagingIndustries

Requirements

Candidates should possess an MS or PhD in Computer Science, Electrical Engineering, Machine Learning, or a related field, or equivalent practical experience, with at least two years of post-degree experience building models in ML Speech technologies such as Automated Speech Recognition/Speech-to-Text (ASR/STT), or Text-to-Speech (TTS). Strong experience working with transformer-based speech models such as Whisper is required, along with proficiency in Python and deep learning frameworks like PyTorch. Familiarity with open-source libraries such as ESPnet, HuggingFace Transformers, torchaudio, or Kaldi is also necessary.

Responsibilities

The Senior ML Research Scientist will develop advanced speech processing pipelines and quality assurance systems for medical dictation workflows, research offline speech processing capabilities and edge computing solutions for clinical environments, and tackle complex challenges unique to healthcare and radiology, such as converting dictated speech into structured data, enabling real-time clinical decision support, or building voice interfaces. They will also work closely with vendor relationships to enhance integration and performance metrics, drive Applied Research by staying current with cutting-edge research and prototyping ideas, and potentially contribute to deploying speech or audio models in production settings.

Skills

Speech Modeling

Machine Learning

Speech Recognition (ASR)

Deep Learning

Natural Language Processing

Healthcare AI

Rad AI

AI-driven software for radiology workflows

About Rad AI

Rad AI enhances radiology workflows using artificial intelligence to improve efficiency and accuracy in radiological practices. Its main product, Omni Reporting, automates routine tasks, ensures follow-up on incidental findings, and improves reporting accuracy. This software integrates seamlessly into existing workflows, making it easier for radiologists to manage their tasks. Unlike competitors, Rad AI emphasizes data security and patient privacy, being SOC 2 Type II and HIPAA certified. The company's goal is to provide reliable AI-driven solutions that streamline healthcare processes and improve patient outcomes.

San Francisco, CaliforniaHeadquarters

2018Year Founded

$76.8MTotal Funding

SERIES_BCompany Stage

AI & Machine Learning, HealthcareIndustries

51-200Employees

Benefits

Health Insurance

Health Savings Account/Flexible Spending Account

401(k) Retirement Plan

Paid Holidays

Remote Work Options

Unlimited Paid Time Off

Risks

Emerging competition from companies like DeepMind could overshadow Rad AI's offerings.

Rapid AI technology evolution requires Rad AI to continuously innovate.

AI-driven automation may face resistance from the medical community.

Differentiation

Rad AI's Omni Reporting won 'Best New Radiology Software' by AuntMinnie.

Rad AI integrates AI with FHIRcast for enhanced radiology workflow interoperability.

Rad AI is a pioneer in using large language models for radiology report generation.

Upsides

Rad AI achieved a 48% increase in radiograph reporting efficiency at RANT.

Rad AI raised $50M in Series B funding, boosting its expansion capabilities.

Strategic collaboration with AGFA HealthCare enhances Rad AI's market position.

Land your dream remote job 3x faster with AI

Try Jobo Free