Multimodal AI Engineer: Audio, Video & Speech

Meta; Denver (CO); On-site; USD 154,000 - 217,000; Full time

Job summary

Meta is seeking an experienced Software Engineer for the Multimedia & Multimodal AI team within Applied AI (AAI). You will own data production, build scalable pipelines across image, video, audio and speech modalities, and design evaluation infrastructure that ties objective metrics to human assessments.

You will mentor engineers, contribute to hiring, and raise quality practices while advancing state-of-the-art multimodal AI capabilities across research, model-training, and engineering partners.

Qualifications

• 3+ years building ML systems in production or research settings.
• Strong Python and PyTorch.
• Demonstrated experience with speech, audio, or music ML.
• Experience with large-scale data pipelines and distributed training.
• Track record of translating research ideas into working, measurable systems.

Responsibilities

• Design and build agentic workflows and pipelines, including human-in-the-loop designs.
• Design and own data pipelines at scale: ingestion, filtering, pseudo-labeling and captioning with attribute classifiers, and provenance tracking for audio corpora.
• Build evaluation infrastructure: objective metrics, human listening-test pipelines, and correlation analysis.
• Improve training efficiency and reliability — distributed training, GPU utilization, codec and tokenizer retraining, experiment management.
• Reproduce and extend state-of-the-art research: implement new methods from papers into our codebases and run ablations.
• Mentor engineers on the team and contribute to hiring and onboarding.
• Build and train generative and representation models for speech, sound, and music.

Skills

Python

PyTorch

Speech/Audio ML

Distributed training

Data pipelines

Research to production

ML systems

Education

Bachelor's degree in CS/EE

Tools

Git

Linux

Job description

Meta is seeking an experienced Software Engineer for the Multimedia & Multimodal AI team within Applied AI (AAI). You will own data production, build scalable pipelines across image, video, audio and speech modalities, and design evaluation infrastructure that ties objective metrics to human assessments.

You will mentor engineers, contribute to hiring, and raise quality practices while advancing state-of-the-art multimodal AI capabilities across research, model-training, and engineering partners.