Founding Engineer

We’re looking for a Founding Engineer to lead core development of our personal agentic AI assistant (Voice OS). This is a hands-on, zero-to-one role, working closely with the founder to design and implement:

• Local + cloud LLM pipelines optimized for real-time, low-latency voice interaction

• Agentic frameworks (tool-calling, memory, contextual orchestration) that make the AI proactive, reliable, and human-aligned

• Voice-first UX layers (STT, TTS, embeddings, RAG pipelines) that feel natural and effortless

• Cross-platform infrastructure starting with Android, expanding to Windows, iOS, and beyond

This is a co-creation role — you’ll experiment, break things, and push limits.

⸻

What You’ll Do

• Architect and implement our Voice OS MVP (real-time conversation, task automation, tool integrations)

• Run LLMs (3B–7B scale) on edge hardware (Snapdragon, Tensor, Apple Silicon) and optimize inference speed

• Build a hybrid local + cloud orchestration framework with LangChain, DSPy, AgentScope, or equivalent

• Integrate core AI tools: summarization, RAG QA, email/calendar automation, note-taking, Spotify/Notion APIs, etc.

• Collaborate with design & hardware teams to create a seamless ear-wearable experience

Preferred Skills & Experience

• 3–5+ years in AI/ML engineering, systems architecture, or voice tech (startup/0→1 preferred)

• Strong knowledge of Python, C++, or Rust (plus: mobile app integration Flutter/React Native)

• Experience with LLM orchestration (tool-calling, memory, multi-agent setups)

• Edge deployment experience with Android (JNI, NDK) or embedded AI frameworks a big plus

• Prior work in voice assistants, wearable tech, or human-computer interaction is highly valued