Software Engineer, Edge AI Systems

About the Role

Join a newly formed team developing and deploying AI systems for operation on tactical edge hardware. You will build specialized models and agents that connect with sensors, effectors, and drone systems, supporting demanding defense-sector use cases where reliable operation on constrained hardware matters.

What You'll Do

• Train and fine-tune language models for edge deployment and tactical applications.
• Optimize inference and runtime performance across GPU, CPU, and specialized accelerator platforms.
• Design and implement multi-agent systems for varied mission scenarios.
• Integrate AI agents with sensors, effectors, and uncrewed vehicle systems.
• Benchmark agent capabilities and evaluate performance, reliability, and safety.
• Build and maintain robust, scalable systems for high-stakes environments.

What We're Looking For

• Four to ten years of software engineering experience, with hands-on experience building and optimizing production LLM or AI-agent systems.
• Experience with model training or fine-tuning, agent optimization, or inference optimization, using tools such as PyTorch, Hugging Face Transformers, LangChain, or vLLM.
• Strong proficiency in Python, C++, and/or Rust, plus experience optimizing software for edge hardware. Familiarity with tools such as Ollama, Mojo, TVM, or TensorFlow Lite is useful.
• A bachelor's degree or higher in computer science or a quantitative field such as physics or mathematics.
• Experience building mission-critical software with rigorous reliability, security, performance, and testing requirements.
• Background in defense, aerospace, robotics, drones, applied research, or other technically demanding environments is valuable.
• Cross-functional or customer-facing experience, US citizenship, and eligibility and willingness to obtain a US security clearance.
• Ability to work from Seattle or relocate there, and travel up to 25%.

Compensation & Benefits

Base salary: $145,000 to $275,000 USD annually. Visa sponsorship is not available.

Location

Hybrid, based in Palo Alto, California.