Imagine WWE SmackDown, but with cute, AI-trained robots. Eight teams spend days training reinforcement-learning policies for Pollen’s open-source MicroDuck, then we throw them in the simulator to fight for the Golden Beak Belt. The whole competition will be livestreamed. The catch: you’re not building a robot, you’re training its brain. Want in? Register a team, train your policy, or host a showdown in your city. When the real MicroDucks ship, we go IRL.
HuggingFace and Pollen Robotics caught the world by surprise by unveiling a new robot called Microduck.
Within days, they had pre-sold more than 10,000 units of the biped.
It could be just another "Kickstarter" project, where you put money in and hope it arrives on time. Instead, the team shipped a simulator and a full training environment so you can already start teaching your robot some moves before it ships.
We've seen people teaching their virtual duck how to break dance or dance like a swan. But what caught my attention was JP's simulation, where a model was trained to let robots play sumo against each other.
This discovery led me to launch DuckFightClub, where robot aficionados meet to compare their training skills in the simulator and eventually in real life.
Do you have what it takes to train a model to play duck sumo, beat them all, and win the Golden Beak Belt?
The first event will be livestreamed on Youtube and some local IRL are already scheduled.
First rule of DuckFightClub: don't quack about it.
@picsoung training a duck to fight is not what i expected to see today 😂 curious what’s been the most unexpected strategy someone has tried so far?
This is the most fun use of reinforcement learning I've seen on here in a while - turning an RL leaderboard into an actual spectator sport is a genuinely good idea. Are the policies trained from scratch each competition or can teams carry over and fine-tune a duck across seasons?
livestreaming a competition whose first rule is not talking about it is a bold read of the source material.
“Train your MicroDuck and win the Golden Beak Belt”
DuckFightClub launched on Product Hunt on September 9th, 2026 and earned 141 upvotes and 13 comments, placing #8 on the daily leaderboard. Imagine WWE SmackDown, but with cute, AI-trained robots. Eight teams spend days training reinforcement-learning policies for Pollen’s open-source MicroDuck, then we throw them in the simulator to fight for the Golden Beak Belt. The whole competition will be livestreamed. The catch: you’re not building a robot, you’re training its brain. Want in? Register a team, train your policy, or host a showdown in your city. When the real MicroDucks ship, we go IRL.
DuckFightClub was featured in Robots (10.7k followers), Sports (11.4k followers) and Artificial Intelligence (478.1k followers) on Product Hunt. Together, these topics include over 130k products, making this a competitive space to launch in.
Who hunted DuckFightClub?
DuckFightClub was hunted by Nicolas Grenié. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how DuckFightClub stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Nicolas Grenié
HuggingFace and Pollen Robotics caught the world by surprise by unveiling a new robot called Microduck.
Within days, they had pre-sold more than 10,000 units of the biped.
It could be just another "Kickstarter" project, where you put money in and hope it arrives on time.
Instead, the team shipped a simulator and a full training environment so you can already start teaching your robot some moves before it ships.
We've seen people teaching their virtual duck how to break dance or dance like a swan. But what caught my attention was JP's simulation, where a model was trained to let robots play sumo against each other.
This discovery led me to launch DuckFightClub, where robot aficionados meet to compare their training skills in the simulator and eventually in real life.
Do you have what it takes to train a model to play duck sumo, beat them all, and win the Golden Beak Belt?
The first event will be livestreamed on Youtube and some local IRL are already scheduled.
First rule of DuckFightClub: don't quack about it.