Sensei GTM brain

Sensei supplies human demonstration training data to robotics companies through a distributed operator network.

category

Robotics training data collection

market profile

  • text: Robotics teams need real-world manipulation demonstrations to train and improve learned robot behaviors.
  • text: Collection vendors compete to deliver human and robot demonstrations with appropriate task coverage and quality.

personas

  • title: Head of robot learning or embodied AI research
    pains:
    • Policies lack sufficient diverse demonstrations for real-world manipulation
    • Researchers spend scarce time supervising collection rather than training models
  • title: Robotics data operations lead
    pains:
    • Teleoperation collection requires hardware, operators and repeatable quality controls
    • Increasing task diversity and collection volume strains internal operations
  • title: Robotics startup CTO
    pains:
    • Existing collection methods can be slow and expensive for an early team
    • Model progress depends on collecting examples across more environments

icp

  • text: Robotics developers training manipulation policies or embodied foundation models from human demonstrations.
  • text: Robotics labs and startups that need outsourced collection across everyday environments and custom tasks.

message pillars

  • text: Scale the supply of robot training examples.
  • text: Access a distributed human collection network.
  • text: Collect examples from real-world environments.

objections

  • objection: Can the data transfer to our robot and task?
    response: Define required sensor modalities and task success criteria, then validate a small sample against the target policy.
  • objection: How do we trust distributed operators to follow the task specification?
    response: Agree on a collection protocol and quality review with Sensei before scaling the request.
  • objection: We can collect data ourselves.
    response: Compare delivered usable demonstrations and time to collection against the cost of operating an internal fleet.

competitors

  • name: Scale AI
    why: Provides physical AI data collection and human demonstration programs.
  • name: PrismaX
    why: Offers teleoperation and egocentric demonstrations through an operator network.
  • name: Appen
    why: Provides custom robotics demonstration and egocentric data collection.

influencers

  • name: Anubhav Guha
    url: https://www.ycombinator.com/companies/sensei
    why: Sensei founder and CEO listed in the YC company profile.

keywords

  • robotics training data
  • robot demonstration data collection
  • robot imitation learning data
  • human demonstration dataset
  • robot manipulation training data
  • robot teleoperation data

voice

  • concise
  • technical
  • mission-oriented

signals

  • text: A robotics lab is building manipulation models and looking for demonstrations beyond its own facilities.
  • text: A team is expanding the number of household or industrial tasks its robot policy must handle.

pricing

model: unknown

motion

sales_led

channels

  • YC launch
  • founder-led sales
  • operator recruitment

confidence

medium

Founders

  • name: Anubhav Guha
    title: Founder/CEO
    role: CEO
    bio: CEO at Sensei - building the marketplace for robotic training data. Previously MIT Robotics PhD (dropped out) and Chief Engineer & Program Manager of Aurora Flight Science/Boeing's DARPA AlphaDogfight team. I've spent my whole career in robotics, and my goal is to help make AI robotics a reality.
  • name: John Piotti
    title: Founder/CTO
    role: CTO
    bio: CTO at Sensei - scaling robotic training data collection to be 2x faster and 10x cheaper. Previously MIT '19, Robotics Engineer at Aurora Flight Science/Boeing, and full stack web development at UtilityAPI. Also building a house in my free time.

Back to YC companies