AI Research Engineer (On-Device ML Focus) | Bay Area | Stealth Startup
We are working with a well-funded, early-stage AI company building something genuinely different at the intersection of agent systems + on-device intelligence.
They’re tackling a very real problem - replacing manual IT workflows (tickets, logs, reactive support) with an AI-native system that can detect, diagnose, and resolve issues autonomously on every device.
What makes this role interesting
This is not just another LLM/backend role.
A big part of the challenge is building lightweight ML systems that run directly on-device (Windows, Mac, Linux), under real-world constraints:
- Limited compute + memory
- Real-time inference requirements
- High-volume telemetry / behavioural signals
- Need for reliability in production (not sandbox experiments)
You’ll be working across
- On-device models (edge ML, optimisation, inference)
- Backend reasoning systems (LLMs, agents, workflows)
- End-to-end pipelines (training → evaluation → deployment)
What they’re looking for
Strong preference for engineers/researchers who have:
- Hands-on experience with on-device / edge ML
- Built or deployed models in constrained environments (CPU, mobile, embedded, etc.)
- Worked on real-world systems, not just research prototypes
- Comfort operating in 0→1 environments with high ownership
Nice to have
- Multimodal systems (vision, audio, sensor data)
- Agent systems / LLM-based workflows
- Experience bridging research → production
Why this team
- Led by a serial founder with multiple successful exits
- Very small team → high ownership + real impact
