🛠 Tech Stack

💼 About This Role

You'll contribute to Zyphra's Audio Team, building open-source audio models like autoencoders and speech-to-speech systems. You'll work on large-scale training runs and architecture improvements. This role offers a chance to publish research in a fast-paced AI company.

🎯 What You'll Do

• Design and train novel audio autoencoder architectures
• Optimize performance of large-scale training pipelines
• Collect and process audio datasets for model training
• Run ablations to improve training methodologies

📋 Requirements

• Strong research taste and ability to execute projects from conception to write-up
• Strong implementation ability in PyTorch and Python
• Expertise in audio models such as TTS, ASR, or speech-to-speech
• Experience with large-scale GPU cluster training

✨ Nice to Have

• Experience with diffusion models or GANs
• Published research in machine learning venues
• Postgraduate degree in a scientific subject

🎁 Benefits & Perks

• 🏥 Comprehensive medical, dental, vision, FSA
• 💰 Competitive compensation and 401(k) plan
• ✈️ Relocation and immigration support
• 🍕 In-office snacks and meals
• 🏖️ Unlimited PTO and company holidays

📨 Hiring Process

Estimated timeline: 2-4 weeks · AI estimate

1. 1Recruiter Call· 30 min
2. 2Technical Interview· 60 min
3. 3Onsite Interview· 4 hours

🚩 Heads Up

• Requirements list is lengthy with many preferred qualifications included
• Job posting mentions 'unlimited PTO' without specifics
• Role may require deep expertise across multiple areas (audio, ML, engineering)