Biologics Stability Scientist: End-to-End Program Lead

Role
Port and optimize a containerized video analytics pipeline to run on CPU-constrained router hardware (Cradlepoint OS, Wi-Fi 7 PrplOS, FWA routers). You'll own the full stack: model optimization, container architecture, and on-device inference performance.
--
- What You'll Do
- Port GPU-based video analytics models (object detection, classification) to CPU-only router targets
- Optimize inference pipeline to stay under 100MB memory footprint using SLMs
- Build containerized architecture with dynamic cloud-driven model loading
- Tune accuracy/performance tradeoffs on ARM/MIPS router hardware
- Integrate with Cradlepoint OS and PrplOS environments
- Benchmark and iterate on detection accuracy vs. latency on constrained hardware
--
- Required
- 4+ years in embedded systems or edge ML deployment

- Experience with containerization (Docker, LXC) on constrained devices
- ML model optimization: quantization, pruning, ONNX, Tensor Flow Lite, OpenVINO
- Video analytics / computer vision (YOLO variants, object detection pipelines)
- Python + C/C++ on Linux embedded targets
- Cross-compilation, profiling, and memory optimization
--
- Strong Plus
- Cradlepoint Net Cloud / PrplOS / OpenWRT experience
- NPU/DSP acceleration on router-class So Cs
- Deep Stream or similar inference pipeline experience (GPU→CPU migration)
- SLM deployment (sub-1B parameter models on edge)
- RTSP/video streaming on embedded Linux
--
- You Are
- Comfortable with no GPU — CPU-only inference is the constraint, not a fallback
- Pragmatic about accuracy tradeoffs at the edge
- Experienced navigating vendor OS lock-in and limited debugging tool chains