Founding AI Engineer

Role: Founding AI Engineer (Multimodal & Real-Time Systems)

About this Role:
We are hiring a unique 'Founding' engineering hire to take full ownership of the initial technical build—from architecture and model integration to deploying a functional consumer product. You will begin as our sole engineer, working directly with founders on an early-stage venture focused on multimodal intelligence.

What you own:

• MVP & Architecture: Translate vision into achievable milestones, build backend systems and mobile-oriented interfaces, and define technical architecture for real-time AI services.
• Multimodal Integration: Engineer capabilities in image/video understanding, voice input/response (WebRTC/WebSockets), contextual reasoning, and session state management.
• Visual Responses & Orchestration: Develop structured visual outputs (text + imagery) and build reliable orchestration for model calls with streaming synchronization to handle user interruptions seamlessly.
• Evaluation & Controls: Establish evaluation datasets, regression checks, monitoring tools, and implement strict controls for camera/microphone access and personal data privacy.

What we need from you:

We seek evidence of shipping, not just tutorials. You must have built deployed AI-enabled software with a strong understanding of asynchronous services, probabilistic system limitations, and operational responsibility.

Your expertise in real-time voice (WebRTC), vision-language models (VLMs), dynamic interfaces, personalization/memory systems, or mobile applications is highly relevant. We value practical experience over exact titles; if you have 4–10 years of hands-on AI product work with a portfolio demonstrating reliability and cost-efficiency, we want to hear from you.

Compensation & Growth:

• Closed for discussion based on demonstrated capability ('You Prove, You Decide').
• Ownership of the initial architecture and direct founder access.
• The opportunity to establish engineering foundations for a new AI business.

How to Apply:

Submit your résumé plus links to 1-2 projects. Explain what you personally designed, one difficult problem solved, and any measured results (latency, cost, reliability). If work is confidential, describe your contribution privately.