AI Researcher (Multimodal Models)

TL;DR

AI Researcher (Multimodal Models): Conducting research on foundational multimodal models for conversational avatars with an accent on video, audio, and language sequence modeling. Focus on building and optimizing architectures like V-JEPA, autoregressive, and diffusion paradigms to advance human-computer interaction.

Location: Hybrid in San Francisco, New York, or London, with remote options within the U.S. or Europe.

Company

hirify.global is a research lab building AI Humans to bridge the gap between people and machines through real-time conversational simulation models.

What you will do

• Conduct research on foundational multimodal models for conversational avatars and neural talking-heads.
• Model video, audio, and language sequences using autoregressive, predictive architectures, and diffusion paradigms.
• Focus on temporal and sequential data processing rather than static imagery.
• Collaborate with the Applied ML team to transition research into production systems.
• Define the cutting edge of multimodal learning by staying current with state-of-the-art research.

Requirements

• PhD or equivalent hands-on research experience in a relevant field.
• Deep understanding of sequence modeling in video, audio, or language domains.
• Expertise in modeling human behavior, affect, or generation.
• Strong background in deep learning, including transformers and diffusion models.
• Proven ability to write production-grade code, specifically in PyTorch.
• Must be located in the U.S. or Europe for remote arrangements.

Nice to have

• Publications in top-tier conferences like CVPR, ICCV, NeurIPS, ECCV, or ACMMM.
• Experience with large model training, including LLMs or VLMs.
• Familiarity with software engineering best practices.

Culture & Benefits

• Flexible work schedule and environment.
• Unlimited PTO policy.
• Competitive healthcare and gear stipends.
• Opportunity to work in a diverse, supportive research lab environment.