Drimva

Aktoria Robotics

Robot Learning Researcher (VLA / Foundation Models)

$450K–$600K

San Francisco, CA, US · Full-time · Posted Oct 29, 2026

About the job

Trains and fine-tunes vision-language-action and robot foundation models, evaluates data and scaling, and publishes research and benchmarks.

What you'll do

  • Train and fine-tune VLAs and robot foundation models.
  • Run data-quality ablations and scaling studies; curate and filter data, define formats, and evaluate on real robots.
  • Publish papers and benchmarks.

What they ask for

  • Experience training IL/VLA policies at scale.
  • First-author publications at CoRL, ICRA, IROS, RSS, or NeurIPS, or equivalent strong open-source work.

Pay

$450K - $600K (base salary)

AI focus

Training and fine-tuning VLAs and robot foundation models, including policy training, data evaluation, and real-robot evaluation.

Model-written summary · the original posting is authoritative

First seen Oct 29, 2026 · Last verified Nov 6, 2026 (UTC)