Axonnic GTM brain

category

market profile

• text: Dexterous robot learning requires examples of physical tasks and information about contact or effort that video alone may not capture.
• text: Physical AI data suppliers use varied capture methods, including wearable gloves, video and teleoperation.

personas

• title: Manufacturing automation director
  pains:

  • Delicate manual tasks resist conventional automation because worker skill is difficult to specify.
  • Capturing training demonstrations must fit around production and worker safety.
• title: Robotics machine learning lead
  pains:

  • Video alone misses muscle activation and inferred effort during manipulation.
  • Demonstrations must be aligned and retargeted to a robot's action space.
• title: Robotics data acquisition lead
  pains:

  • Gloves can obstruct realistic hand work while occlusion complicates camera-only capture.
  • Gathering task-specific multimodal data and validating it is resource intensive.

icp

• text: Manufacturers with repetitive, high-value dexterous manual processes and an automation pilot budget.
• text: Robotics and embodied AI labs training manipulation policies from task-specific human demonstrations.
• text: Robot integrators partnering with factories on manipulation tasks where effort or contact information matters.

message pillars

• text: Teach robots using the skill of expert workers.
• text: Capture information beyond what video alone reveals.
• text: Let people demonstrate with their hands unobstructed.

objections

• objection: Does EMG actually improve robot performance for our task?
  response: Run a controlled task-specific pilot comparing success rates against a baseline without EMG.
• objection: Will workers have to wear a sleeve in production?
  response: The described workflow uses the sleeve during demonstration; the deployed robot uses its own sensors.
• objection: Will capture disrupt our line?
  response: Agree on a representative task, capture schedule and worker participation before the pilot.

competitors

• name: Mecka
  website: https://www.mecka.ai/
  why: Collects human activity data and builds physical AI training and deployment infrastructure.
• name: MANUS
  website: https://www.manus-meta.com/use-cases/abb-x-psyonic-human-demonstration-data-for-robot-dexterity
  why: Data gloves capture hand motion for robot demonstration datasets.
• name: Telepath
  website: https://telepathrobotics.ai/robotics-data-collection
  why: Provides task-specific robot training demonstrations using alternative capture methods.

influencers

• name: Maxim Williams
  url: https://uk.linkedin.com/in/maxim-williams
  why: Axonnic founder sharing demonstrations and product launch updates.

keywords

• robot training data
• EMG robot learning
• dexterous manipulation data
• human demonstration robotics
• robot imitation learning
• manufacturing robot dexterity

voice

• Demonstration-led
• Technical and measurable
• Pragmatic about manufacturing

signals

• text: A factory is piloting automation for assembly or handling tasks that need precise finger manipulation.
• text: A robotics lab is collecting multimodal manipulation demonstrations or building a new policy dataset.

stack

• EMG
• video
• hand pose tracking

pricing

motion

channels

• LinkedIn
• YC launch
• product demos
• industry partnerships

confidence

Founders

• name: Maxim Williams
  title: Founder/CEO
  role: CEO
  bio: Founder of Axonnic. Oxford MBiol Biology and author of research on HD-sEMG. Previously built internal tools at Skyscanner and early employee at Voxblock, working across PCB development, supply chain and its Series A fundraise.
• name: Arsh Patankar
  title: Founder/CTO
  role: CTO
  bio: Dev -> Medic -> Founder Oxford 22-26