Ex-Meta Engineer's Dog Accidentally Codes Game by Walking on Keyboard
The threshold for AI programming is being redefined once again as "atmospheric programming" cases continue to multiply. Caleb Leak, a former Meta research engineer, recently shared an amusing experiment: he successfully taught his 9-pound Corgi, "Momo," to develop a fully functional game using Claude Code.

To enable this "canine coding," Caleb devised a precise hardware and algorithmic workflow. Momo provides input by randomly tapping a Bluetooth keyboard connected to a Raspberry Pi 5. This signal is processed by a small program called "DogKeyboard" for filtering before being sent to Claude Code. To motivate Momo to keep "creating," the system automatically dispenses treats from a smart feeder whenever a preset amount of text is entered.
The experiment's core challenge was how the AI could interpret random code. Caleb crafted highly specific prompts for Claude, framing it as an eccentric genius designer who communicates only in extraordinary ways. He informed Claude that the random keystrokes contained a hidden, brilliant game concept. Using automated feedback tools like screenshot analysis, scene validation, and shader verification, Claude successfully transformed these random characters into a playable game within one to two hours.
Momo's latest creation is titled "Quasar Saz." In this game, players assume the role of Zara, battling with a cosmic string instrument. The game was built using the Godot 4.6 engine, with 100% of its logic code written in C#. It features six levels, boss battles, and impressive graphics and sound effects.
Related article
DeepMind CEO Hassabis: I sleep six hours a day, usually feel energetic around 1 a.m.
Fortune recently featured an interview with Demis Hassabis, CEO of Google DeepMind, revealing his unconventional approach to rest and productivity. Hassabis disclosed that he sleeps very little, structuring his waking hours into two distinct work blo
OpenAI, Anthropic Vie for Market Share Despite Revenue Shortfalls
Despite recent reports suggesting OpenAI missed revenue targets, creating pressure on tech stocks this Tuesday, private AI lab investors remain resilient. Seasoned backers have confirmed they will not reduce investment despite negative media coverage
California AV Compliance: A New Era of Tickets, Geofences, and 1M Miles
Guident operates an AuveTech shuttle in South Florida, managing a four-mile route in West Palm Beach and a one-mile route in Boca Raton using its remote monitoring technology. | Credit: GuidentCalifornia is redefining the regulatory landscape for dri
Related Special Topic Recommendations
Comments (0)
0/500
The threshold for AI programming is being redefined once again as "atmospheric programming" cases continue to multiply. Caleb Leak, a former Meta research engineer, recently shared an amusing experiment: he successfully taught his 9-pound Corgi, "Momo," to develop a fully functional game using Claude Code.

To enable this "canine coding," Caleb devised a precise hardware and algorithmic workflow. Momo provides input by randomly tapping a Bluetooth keyboard connected to a Raspberry Pi 5. This signal is processed by a small program called "DogKeyboard" for filtering before being sent to Claude Code. To motivate Momo to keep "creating," the system automatically dispenses treats from a smart feeder whenever a preset amount of text is entered.
The experiment's core challenge was how the AI could interpret random code. Caleb crafted highly specific prompts for Claude, framing it as an eccentric genius designer who communicates only in extraordinary ways. He informed Claude that the random keystrokes contained a hidden, brilliant game concept. Using automated feedback tools like screenshot analysis, scene validation, and shader verification, Claude successfully transformed these random characters into a playable game within one to two hours.
Momo's latest creation is titled "Quasar Saz." In this game, players assume the role of Zara, battling with a cosmic string instrument. The game was built using the Godot 4.6 engine, with 100% of its logic code written in C#. It features six levels, boss battles, and impressive graphics and sound effects.
DeepMind CEO Hassabis: I sleep six hours a day, usually feel energetic around 1 a.m.
Fortune recently featured an interview with Demis Hassabis, CEO of Google DeepMind, revealing his unconventional approach to rest and productivity. Hassabis disclosed that he sleeps very little, structuring his waking hours into two distinct work blo
OpenAI, Anthropic Vie for Market Share Despite Revenue Shortfalls
Despite recent reports suggesting OpenAI missed revenue targets, creating pressure on tech stocks this Tuesday, private AI lab investors remain resilient. Seasoned backers have confirmed they will not reduce investment despite negative media coverage





Home






