Location: Brooklyn / San Francisco (Relocation Supported) Type: Full-time | Competitive Pay + Equity Start Date: ASAP
- Deploy compact, low-latency AI models on edge devices under severe resource constraints
- Engineer real-time inference across heterogeneous environments: ARM, RISC-V, x86, and mobile-class platforms
- Optimize model execution pipelines across CPUs, GPUs, NPUs, custom silicon
- Build embedded runtime systems for degraded, disconnected, and contested environments
- Develop intelligent, performance-aware agents with tight control loops and minimal overhead
- Collaborate closely with the founding team to define our low-SWaP infrastructure and core product direction
- 3+ years in embedded systems or systems-level software
- Strong C/C++, Rust, Python skills
- Experience deploying AI inference on edge hardware (e.g. TensorRT, ONNX Runtime, TVM, Apple Neural Engine, or Metal)
- Familiarity with quantization, pruning, & distillation of models for real-time edge execution
- Knowledge of model compression frameworks (e.g. PyTorch Lightning, Hugging Face Optimum, TensorFlow Model Optimization)
- Experience with real-time OS (e.g. FreeRTOS, Zephyr) and direct hardware access
- Comfort working across memory constraints, volatile compute environments, and secure firmware deployments
- Exposure to edge deployment environments such as macOS (M1/M2), embedded Linux, and secure radios
- Prior work with over-the-air model updates or remote inference in disconnected networks
- Familiarity with DSPs, SDRs, or custom ASICs
- Kernel-bypass networking (DPDK, eBPF) in embedded or mobile contexts
- Background in national defense systems, aerospace, or robotics
- Strong understanding of secure boot, firmware verification, or TPM-backed deployments
We’re rethinking how machines communicate—from the edge up. Today's networks weren’t designed for autonomous agents, disconnected inference, and high-performance coordination. It’s a ground-up redesign of how data moves. If you’re obsessed with performance, control, and building intelligent software close to the hardware, we want to meet you.
Email the following to team@cosmiclabs.io
:
Subject line: Founding Embedded Engineer / [Your Name]
In the body:
- Your name
- Why this role and why Cosmic Labs
- What you bring technically
- Soonest available start date
- Education and years of experience
- Work eligibility status
Attachment:
- PDF resume
Deadline: Applications reviewed on a rolling basis. Final deadline is EOD August 28, 2025.
Let me know if you want a compact version for a GitHub repo index (README.md
) or site footer.