Applied Researcher - Deployment Intelligence & Continuous Learning

Dyna Robotics
Redwood City
Workplace: OnsiteFull timeFunction: Research & Scientific (R&D)Skills: ["Communication","Bias to ship","Problem-solving"]

Own the research-to-production loop for a fleet of real-world robots. Mine high-frequency multimodal sensor and video data to identify failure modes, drift, and regressions, then design monitoring and continuous learning pipelines. Apply reinforcement learning (including offline RL and RL fine-tuning) to improve policies from deployment data. Partner with Research, Data, and Deployment teams to ship measurable model improvements.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Dyna Robotics
Dyna Robotics
1 month ago

Applied Researcher - Deployment Intelligence & Continuous Learning

âś“ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 10 hours agoStatus: Live

Job Summary

Own the research-to-production loop for a fleet of real-world robots. Mine high-frequency multimodal sensor and video data to identify failure modes, drift, and regressions, then design monitoring and continuous learning pipelines. Apply reinforcement learning (including offline RL and RL fine-tuning) to improve policies from deployment data. Partner with Research, Data, and Deployment teams to ship measurable model improvements.
Location: Redwood City
Workplace: Onsite
Employment Type: Full time
Job Function: Research & Scientific (R&D)

Key Responsibilities

  • •Design and ship continuous learning pipelines that turn deployment data (successes, failures, teleop corrections) into targeted fine-tuning and online policy improvement.
  • •Mine high-frequency multimodal sensor and video data across large numbers of fleet episodes to detect failure modes, drift, and regressions.
  • •Apply reinforcement learning methods (offline RL, RL fine-tuning, reward modeling from human and teleop feedback) to improve policies from real-world deployment data.
  • •Build automated fleet monitoring to flag anomalies, near-failures, and out-of-distribution scenes in real time, deciding what needs human review.
  • •Create evaluation harnesses and data-selection strategies to characterize and close cross-scene generalization gaps as robots move to new sites.

Key Requirements

  • •Hands-on experience in at least two of reinforcement learning, sensor-data modeling/anomaly detection, vision-language models, or continual/online learning.
  • •Experience building monitoring, evaluation, or data pipelines for a live ML system using real-world data (not just curated benchmarks).
  • •Comfort designing and reading production experiments (A/B tests, canary rollouts, staged fleet deployments) with statistical rigor.
  • •Strong Python and PyTorch (or JAX), comfortable with large multimodal datasets and distributed compute (Slurm/GPU clusters).
  • •Bachelors, Masters, or PhD in CS, Robotics, Statistics, or a related field (or equivalent practical experience).
Experience:RoboticsReinforcement learningComputer vision
Education:
Skills:CommunicationBias to shipProblem-solving
Tech Stack:PythonPyTorchJAXSlurmGPU clustersReinforcement learningOffline RLReward modelingAnomaly detectionMultimodal sensor dataVision-language models

Company Brief

Dyna Robotics
Develops embodied-AI powered robotic arms and foundation models (DYNA-1) to automate repetitive, stationary tasks across hotels, restaurants, laundromats and retail. Focuses on affordable, deployable robots that generalize across environments and improve with on‑device learning.
Industry: Robotics
Company Size: Small (11 to 50 employees)
Growth: Growth Stage Startup
Valuation: USD 500M to 1B
Funding: Series A
Headquarters: Redwood City, United States
Founded: 2024
WebsiteLinkedInGlassdoor