About the Role
Join our speech pillar to drive coverage of the next AI interface, including text-to-speech, speech-to-text, voice cloning, and real-time voice agents. You will build and extend speech evaluations and arenas, work at a deep technical level with leading speech AI companies, and shape how the industry measures voice technology.
Responsibilities
- Benchmark the speech frontier across text-to-speech, speech-to-text, and speech-to-speech models
- Design and extend evaluation frameworks, prompt libraries, and arenas reflecting real-world developer usage
- Partner with top speech AI companies to benchmark models and shape industry measurement standards
- Produce influential leaderboards, reports, and analysis on speech AI progress
- Drive the product roadmap for the speech benchmarking platform
- Utilize AI-native workflows and cutting-edge tools to maintain a competitive edge
Requirements
- 3+ years of professional experience
- At least 1 year of hands-on experience working with speech AI
- Strong analytical and critical thinking skills
- Proficiency in Python and data analysis
- Hands-on familiarity with modern speech models and evaluation methods including quality assessment, word error rate, latency measurement, and preference testing
- Demonstrable interest and informed opinions regarding the future of Frontier AI
Skills
- Python
- Data analysis
- Speech AI evaluation
- Quality assessment
- Word error rate measurement
- Latency measurement
- Preference testing
Location
- San Francisco, CA
Work Type
- On-site
Experience Level
- 3+ years of professional experience
Salary/Compensations
- Competitive compensation including equity
Benefits
- Equity
- Opportunity to shape AI development priorities
- Exposure to frontier AI models and industry leaders
About the Company
- Leading independent AI benchmarking company
- Trusted by major AI labs including OpenAI, Google, Meta, NVIDIA, and Anthropic
- Team of 40+ employees with plans to double by year-end
- Backed by prominent industry investors including Nat Friedman, Daniel Gross, Andrew Ng, Adam D’Angelo, and Clem Delangue
