Neural Network Specialist ZR TEST

At Special Projects Engineering, we rethink problems from the ground up. We’re a small, high-impact team based in Seattle focused on applying advanced AI to complex, often unconventional challenges. Our work spans autonomous systems, predictive infrastructure, synthetic biology, and defense-grade automation. If it’s ambitious, technical, and hasn’t been done before, we’re probably already working on it.

As a Neural Network Specialist
, you’ll be responsible for designing and optimizing deep learning architectures that push the limits of what’s technically possible. You’ll collaborate closely with the CTO and a diverse team of engineers, scientists, and domain experts to move ideas from prototypes to deployment.

• Design, train, and evaluate neural network architectures for real-world use

• Work with unstructured data including visual, auditory, textual, and time-series formats

• Develop systems for tasks like perception, decision‑making, prediction, and control

• Optimize models for performance, scalability, and deployment on various platforms

• Rapidly iterate on ideas while balancing reliability and experimental curiosity

• Collaborate with hardware, embedded, and software teams to integrate models into real systems

• Contribute to the ML roadmap, infrastructure, and toolchain

• 5+ years of experience in deep learning or applied neural network research

• Strong Python skills, with additional experience in C++ or Rust preferred

• Expertise in PyTorch, Tensor Flow, or JAX

• Solid understanding of training techniques, loss functions, evaluation metrics, and deployment strategies

• Experience with one or more domains like computer vision, NLP, reinforcement learning, or multimodal models

• Comfortable owning projects from research through production

• Bonus if you’ve worked with robotics, embedded AI, or unusual sensor inputs

• Competitive salary and meaningful equity

• Remote‑friendly with flexible hours

• Full medical, dental, and vision coverage

• Access to top-tier GPU hardware

• Small team, zero bureaucracy, high ownership

• Projects that are genuinely interesting and occasionally a little insane