A new frontier
for AI voice

About the role:

Explore and ship low-latency speech systems for AI players that need to talk while playing games in real time. You will work on duplex and streaming voice models, speech understanding, turn-taking, interruption handling, prosody/emotion, latency/quality tradeoffs, and conversational evals.

Your Canvas:

• Prototype duplex and streaming speech model architectures for live gameplay
• Drive speech understanding, turn-taking, interruption handling, and low-latency response loops
• Build evals that measure whether voice feels socially present, not just correct
• Work across model research, inference, data, and product integration

You might be a great fit if you have:

• Deep audio, speech, or voice-modeling experience
• Strong taste around latency, conversational feel, and real-time UX
• Experience shipping applied ML systems beyond notebook prototypes
• Ready to work in-person in San Francisco, CA

What We Offer:

• $150K-$200K salary based on experience
• 1.0-2.0% equity
• Benefits include company-sponsored medical, dental, & vision

About the role:

Own the full stack for how AI companions play games: game-state understanding, learned policies/controllers, world models, action representation, MLP/policy heads, memory, evals, data pipelines, inference, and product iteration.

Your Canvas:

• Build models and control systems that understand game state and decide what to do next
• Design action representations, policy heads, world-model experiments, and gameplay evals
• Own data pipelines, training/fine-tuning, inference, memory, and production behavior
• Turn messy research prototypes into live gameplay that feels competent and responsive

You might be a great fit if you have:

• Shipped applied ML systems and owned model behavior in production
• Experience with agents, RL/imitation learning, multimodal models, or game AI
• Strong engineering range across modeling, data, infrastructure, and product loops
• Ready to work in-person in San Francisco, CA

What We Offer:

• $150K-$200K salary based on experience
• 1.0-2.0% equity
• Benefits include company-sponsored medical, dental, & vision

About the role:

As our founding Product Engineer, you’ll be building an AI companion app and directly interfacing with users. Imagine an AI companion that you can raise like a pet, play games with, and even bring it across other app contexts and webpages.

This is a full-spectrum role. You'll write the code, design the app, talk to the users, and help bring it to market. We're small, the surface area is enormous, and we’re looking for candidates who are comfortable owning large projects and capable of performing fast user-facing experiments.

What you'll build:

• Design and develop a real-time voice companion app across iOS, Android, and desktop.
• Build the systems behind low-latency voice, character animation, and shared-activity experiences (playing and watching together)
• Own features from idea to launch: prototype it, ship it, watch the metrics, talk to the users, iterate
• Help take the product to market - this includes Tiktok shorts, Discord messages, Youtube videos, and more.

You might be a great fit if you:

• Have experience owning the entire stack for iOS/Android/Desktop
• Build games and are familiar with implementing gamification features
• Are familiar with vibe-coding tools like Claude Code, Codex, Cursor, etc.
• Love talking to users and interpreting their feedback
• Have posted a video to social media that’s gone viral.
• Are ready to work in-person in San Francisco, CA
• Are at least a little weeby

Strong candidates may also have:

• Prior work on AI voice agents
• Experience with game engines, animations, and tech art
• Grown their own audience before for another app

What we offer:

• $100K–$140K salary based on experience
• 0.5-1.0% equity
• Benefits including company-sponsored medical, dental & vision