Software Engineer (Data & Evals)

This onsite opportunity in San Francisco offers you the chance to join a nimble, high-ownership team building data infrastructure for cutting-edge AI systems. You’ll work hands-on with a small group of engineers and researchers, shaping scalable pipelines that handle tens to hundreds of terabytes of multimodal data, with a focus on speech-to-speech research and productionizing models. The role combines impactful development with strong collaboration and clear ownership, backed by competitive compensation and comprehensive benefits.

Benefits

• Competitive base salary of USD 180,000 to 250,000 per year
• Significant equity as an early team member
• Immigration support
• Fully covered medical, dental, and vision insurance
• 401(k) plan
• Onsite collaboration in San Francisco
• Opportunity to work with 100+ TBs of data and large-scale AI models

Responsibilities

• Build and scale infrastructure and distributed data pipelines for large-scale AI and ML systems
• Process and manage tens to hundreds of terabytes of multimodal data
• Support data systems used for training, evaluation, and improvement of speech-to-speech AI models
• Develop batch processing, real-time streaming, and distributed orchestration capabilities
• Design reliable pipelines for speech data transformation, filtering, evaluation, and model improvement
• Collaborate closely with a small, high-performing engineering and research team
• Bridge cutting edge AI research with real-world production environments

Requirements

• Experience building infrastructure and distributed data pipelines to process tens of terabytes of data
• Proven track record working with multimodal data in AI/ML products or systems
• Strong expertise in batch processing, real-time streaming systems, and distributed orchestration
• Hands-on experience with Spark, Kafka, Flyte, Kubernetes, or similar technologies
• Solid software engineering fundamentals and ability to build reliable, scalable systems
• Fast learner who adapts well in a dynamic startup environment
• Strong ownership mindset and ability to work independently with high autonomy
• Comfort working in person with the San Francisco team

Technologies

Key tools include Spark, Kafka, Flyte, and Kubernetes

Nice to Have

• Experience in early-stage startups
• Independent project creation, startup experience, side projects, or open-source work
• Background in transformation pipelines for speech processing
• Experience with transcription, diarization, speech enhancement, filtering, or audio data processing
• Background working with large-scale AI models or ML infrastructure
• Interest in voice AI, speech systems, conversational AI, or multimodal AI products

Interview Process

• 30-minute introductory conversation
• Two technical interviews
• Two culture interviews
• Onsite co-working session collaborating with the team on a data system