about this event
Awign and Armature AI Labs, in partnership with RIC (Robotics India Community) and Tanna TechBiz (NVIDIA Preferred Partner), are hiring interns to work on humanoid robot learning. They are skipping the resume screen and asking you to build something instead.
Record yourself doing a simple task on your phone. Use that footage to make a humanoid robot do it in simulation. The strongest submissions get a 2-month, full-time robot learning internship and full-time roles with the Awign hardware team in Bangalore.
The challenge: Collect a small manipulation dataset yourself. A phone camera is enough, and egocentric footage works well. Then use it to drive a humanoid or dexterous robot hand in simulation. Unitree G1 and MuJoCo are preferred, and alternatives are welcome. There is one hard rule: the data you personally recorded has to play a real role in your approach. Everything else is open.
Here is an example: Film your hand picking up a cup and placing it on a coaster. Retarget your hand poses to the G1's five-finger hand. Post-train a small VLA like SmolVLA on that data. Then show the G1 doing the task in MuJoCo, using fingertip contact to confirm the grasp.
Directions you could take: Post-train a policy on a simple task using your own recordings. Retarget human hand motion to a five-finger hand and use tactile signals for grasp detection or force control. Bootstrap a policy and push its success rate up with RL. Take the world-model route: show video or state prediction, with less focus on policy success. Take a standard policy and make it run much faster than the baseline.
Pick your robot: The Unitree G1 is the default, but other embodiments are welcome. Retargeting your data well to a less common body counts in your favour. All of these have free MuJoCo models in MuJoCo Menagerie: Unitree G1 (default), a 1.3 m humanoid — pair it with a five-finger hand (Inspire, Dex5, or similar); Booster T1, a smaller, lighter humanoid good for fast RL iterations; Apptronik Apollo, a full-size, human-scale humanoid, where human data maps more directly.
Pick your simulator: MuJoCo is preferred. Use whatever runs on the hardware you have. MuJoCo (preferred) is a pip install that runs on a laptop CPU or free Colab; use MJX or MuJoCo Playground for GPU-parallel RL. PyBullet is the lightest option and runs anywhere. Isaac Sim / Isaac Lab is best for large-scale RL and photoreal cameras but needs an NVIDIA RTX GPU. LIBERO / robosuite are tabletop benchmarks built on MuJoCo, good for starting with an arm.
What to submit: A public GitHub repository by Wed 4 Nov 2026, 23:59 IST, containing a README or short deck covering how to run it, example outputs (videos or GIFs), your design choices, and what worked and what didn't; and your recorded data (or a representative sample), plus a note on how you collected it. The submission form will be emailed to approved registrants.
How we judge: Creativity — as long as your own data genuinely matters to the result. Results — how well the robot performs in simulation, and/or the quality of your world-model predictions. Clarity — a simple implementation and an honest, clear presentation; no AI slop please.
Challenge winners: Paid internships with Awign — full-time, 6 days a week, in Bangalore, for 2 months. You will work daily with the founding team and engineers working on humanoid robots. Strong performers will be considered for full-time roles with the Awign hardware team.
About Awign (a Mynavi company): robotics data collection at scale. Armature AI Labs: a shared physical AI and robotics lab in HSR Layout, Bangalore. RIC: Robotics India Community, democratising robotics in India. TTB (Tanna TechBiz): Physical AI Solutions, NVIDIA preferred partner.
Questions? Email hello@armatureailabs.com
tracks
Policy Post-Training
Post-train a policy on a simple task using your own recordings.
Retargeting and Tactile Control
Retarget human hand motion to a five-finger hand and use tactile signals for grasp detection or force control.
RL Bootstrapping
Bootstrap a policy and push its success rate up with reinforcement learning.
World Models
Show video or state prediction, with less focus on policy success.
Speed Benchmarking
Take a standard policy and make it run much faster than the baseline.
when
Humanoid Challenge: Teach a Humanoid With Your Own Hands
Challenge opens; participants record their own manipulation data and build a simulation pipeline.
◍ OnlineSubmission Deadline
Public GitHub repository with README/deck, example outputs and recorded data due by 23:59 IST.
◍ Onlinetickets
| tickets | price | Fee | Status | description |
|---|---|---|---|---|
| Challenge Registration | Free | No fee | Available | Free entry; the submission form is emailed to approved registrants. |
prizes
| # | PRIZE | price |
|---|---|---|
| Winner | Paid internship with Awign — full-time, 6 days a week, in Bangalore, for 2 months, working daily with the founding team and humanoid robot engineers; strong performers considered for full-time roles with the Awign hardware team | TBA |