Build generative AI systems
deployed for clients.
22-person team. 10M+ users. 1T+ tokens/month. We hire people who ship real products, not prototypes.
San Francisco co-founders. Distributed team. Remote-friendly.
Scope
Every engineer touches LLM orchestration, GPU inference, voice and video synthesis, consumer surfaces, payments, and analytics. Often in the same week.
Impact
Your work ships to client products used by millions. Not a research paper or pitch deck — systems people use daily and pay for.
Measurement
Every change ties to a hypothesis. Weekly experiment cycles with real data. You see results in days, not quarters.
Ownership
No layers between you and production. You own systems end-to-end. Small team, high trust, real autonomy.
What you'll work on
LLM Systems
Multi-provider orchestration, dynamic routing, structured context management, RLHF-informed tuning. 1T+ tokens/month.
Generation Infrastructure
Diffusion-based image synthesis, neural video generation, GPU cluster management. Inference optimization for latency, throughput, and cost.
Voice & Audio
Neural TTS, voice cloning, emotional expression modeling, real-time bidirectional audio at conversational latency.
Consumer Product
React / Next.js / TypeScript. Product analytics, A/B testing, conversion optimization, subscription and virtual currency systems.
Platform & Security
High-risk payments, fraud detection, DDoS mitigation, monitoring (Sentry, Grafana), CI/CD, ETL pipelines.