Member of Technical Staff - AI Training Platform

Unconventional AI is developing energy-efficient computing foundations for artificial intelligence by mapping neural networks directly to semiconductor device physics. The Member of Technical Staff, AI Training Platform will build and scale infrastructure for training, evaluating, and benchmarking models, while optimizing software and enabling researchers to evaluate novel hardware.

Responsibilities:

• Training Infrastructure: Architect, scale, and maintain the core AI/ML/RL training platform and infrastructure. Design and scale multi-node distributed training systems, implementing elastic sharding and robust data streaming pipelines for fast, large-scale iteration. Implement and robust model checkpointing and recovery mechanisms
• Framework Development: Maintain and expand our proprietary training framework, ensuring it provides a robust and flexible foundation for all internal model training
• Optimization & Benchmarking: Develop and optimize kernels using low-level programming models like CUDA and Triton. Design rigorous benchmarking suites to track Model Flops Utilization (MFU), memory bandwidth, and convergence stability
• Hardware Evaluation Tooling: Build and iterate on tooling to enable rapid, high-level evaluation of novel hardware ideas against benchmarks
• Cross-Functional Collaboration: Act as a translator, discussing algorithmic trade-offs with theorists and converting model requirements into concrete specifications for infrastructure and hardware engineering teams
• Optimization & ML Hillclimbing: Design systems to track and visualize quality-vs-efficiency Pareto frontiers, enabling researchers to optimize models for both performance and energy consumption