U

Unconventional AI

Analog computing substrate for energy-efficient AI

System Modeling - Performance Models

Hardware Systems IntegrationConfirmed live in the last 24 hours
Full-timeNo salary listedRemote

About the Job

Requirements

  • A Master of Science or PhD in a quantitative field such as AI/ML, Computer Science, Physics, Electrical Engineering, or Applied Mathematics, or a Bachelor's degree with substantial, clear evidence of equivalent research or engineering depth.
  • Experience with tools and development for power profiling, modeling, and simulation for AI workloads.
  • Deep understanding of spatial architectures and data orchestration mechanisms.
  • Deep understanding of dataflow strategies and their tradeoffs, including Weight-Stationary, Output-Stationary, Input-Stationary, and Row-Stationary.
  • Familiarity with open-source tools for hardware accelerator design, including TimLoop, Accelergy, NeuroSim, CIMLoop, and CACTI.
  • Familiarity with existing systolic array accelerator architectures for AI/ML workloads.
  • Solid understanding of modern AI/ML architectures and training/inference workflows.
  • Experience implementing and debugging ML models in PyTorch or similar, with practical experience profiling, optimizing, and stabilizing non-trivial large-scale ML systems.

Responsibilities

  • Build extensible and composable high-fidelity power, performance, and area estimation tools for novel AI acceleration system architectures to enable rapid design-space exploration.
  • Define and create comparative analyses across candidate architectures and existing state-of-the-art implementations.
  • Work with other teams to understand their needs for modeling and simulation to support high-level system design and lower-level hardware verification.

Desired Qualifications

  • Basic familiarity with analog dynamic systems, including transient responses, nonidealities such as nonlinearity, quantization, random noise, and feedback/stability.
  • Strong Python engineering skills, including modular design, testing, packaging, and continuous integration.
  • Experience with PyTorch internals, including autograd, custom modules, and low-level operations; familiarity with torch.compile or similar graph-capture/compile flows.
  • Experience with CUDA, Triton, or other GPU programming approaches, including writing custom kernels, understanding memory hierarchy, and basic performance tuning.
  • Comfort with some of JAX, NumPy, TensorFlow, Modal, high-performance computing patterns such as MPI, NCCL, and distributed training, or SciPy.
  • Ability to reason across algorithm, software, runtime, and hardware layers.
  • Ability to connect model architecture choices to system performance implications, including memory bandwidth, communication patterns, latency, energy, and numerical issues.
  • Experience applying efficiency techniques such as quantization, sparsity, pruning, distillation, or kernel fusion.
  • Prior experience building or extending a substantial simulation or modeling framework in ML systems, physics, circuits, or another technical domain.
  • Comfort with modeling approximations and tradeoffs, including knowing when to use a simple model and when a more physics-based model is needed.

About the Company

Unconventional AI

Analog computing substrate for energy-efficient AI

Unconventional AI designs analog AI hardware and silicon substrates that run neural networks directly in hardware. Its chips use analog circuits that mimic biological neuron dynamics to process AI workloads in the physics of silicon, reducing energy use compared with digital GPUs. It differentiates itself by pursuing an in-hardware, neuromorphic-style substrate rather than optimizing existing GPUs, supported by prominent venture funding. Its goal is to counter the AI energy bottleneck and enable scalable AI in data centers with biology-scale efficiency.

Company size
51-200
Company stage
Seed
Total funding
$475M
Headquarters
San Diego, CA
Founded
2025

Recently Posted Jobs

Explore more opportunities at Unconventional AI

See more jobs