Multimodal Foundation Model Engineer: Pretraining & Scaling

Hark

San Jose (CA)

On-site

USD 180,000 - 450,000

Full time

8 days ago

A complete application in a minute

Tailored resume and cover letter, ready to send.

Get past ATS filters

Job summary

Hark is building the next generation of AI experiences with multimodal models and specialized hardware. The Omni team is responsible for developing multimodal foundation models, spanning data curation, mixture design, model architecture, scaling, distributed training, and evaluation to enable real-time, cross-modal reasoning and response capabilities.

You will research architectures, design experiments, and implement training recipes with data researchers and training engineers to push the

Qualifications

  • Hands-on experience training large models with trade-offs between quality, compute cost, and stability.
  • Strong experimental judgment including controlled experiments to distinguish hypotheses.
  • Experience applying scaling laws to experiment design and extrapolations.
  • Experience implementing and debugging distributed training workloads across GPU clusters.

Responsibilities

  • Research model architecture, optimization, and scaling for large models.
  • Establish baselines and run controlled experiments to evaluate changes at scale.
  • Develop training recipes, including learning-rate schedules, context lengths, and compute allocation.
  • Investigate mixture-of-experts architectures, hybrid attention, and long-context training.
  • Diagnose loss spikes and numerical issues with training engineers.

Skills

Large-model training
Experimental design
Distributed training
Research & engineering

Job description

Hark is building the next generation of AI experiences with multimodal models and specialized hardware. The Omni team is responsible for developing multimodal foundation models, spanning data curation, mixture design, model architecture, scaling, distributed training, and evaluation to enable real-time, cross-modal reasoning and response capabilities.

You will research architectures, design experiments, and implement training recipes with data researchers and training engineers to push the

Get your free, confidential resume review.

or drag and drop your file here.

Similar jobs

Similar jobs worth comparing

Multimodal Data Scientist, Pretraining Specialist
Multimodal Data Scientist, Pretraining Specialist

Hark • San Jose (CA)

On-site
USD 180,000 - 450,000
Senior Staff, Multimodal Pretraining and Scaling
Senior Staff, Multimodal Pretraining and Scaling

Hark • San Jose (CA)

On-site
USD 180,000 - 450,000
Senior Multimodal AI Systems Engineer – San Jose
Senior Multimodal AI Systems Engineer – San Jose

Hark • San Jose (CA)

On-site
USD 180,000 - 450,000
Staff Engineer - Foundation Models & Pretraining
Staff Engineer - Foundation Models & Pretraining

Hark • San Jose (CA)

On-site
USD 180,000 - 450,000
Multimodal ML Research Intern: Omni-Model Innovation
Multimodal ML Research Intern: Omni-Model Innovation

Stfrancis • Palo Alto (CA)

On-site
USD 80,000 - 125,000
Paid sick leave
Up to 13 paid holidays annually
Company-sponsored medical plan (full‑/
Member of Technical Staff, Pretraining Architecture and Scaling
Member of Technical Staff, Pretraining Architecture and Scaling

Hark • San Jose (CA)

On-site
USD 180,000 - 450,000
Member of Technical Staff, Multimodal
Member of Technical Staff, Multimodal

Hark • San Jose (CA)

On-site
USD 180,000 - 450,000
Multimodal AI Research Intern — Omni-Modal Models
Multimodal AI Research Intern — Omni-Modal Models

Lightspeed Studios • Palo Alto (CA)

On-site
USD 80,169 - 120,000
Paid sick leave
Medical plan enrollments
Senior Multimodal Pre-Training Architect
Senior Multimodal Pre-Training Architect

Walden Robotics • San Francisco (CA)

On-site
USD 280,000 - 430,000
Annual cash bonus
Company equity
Company-subsidized insurance programs
+3
Member of Technical Staff, Pretraining
Member of Technical Staff, Pretraining

Hark • San Jose (CA)

On-site
USD 180,000 - 450,000