Research Engineer, Inference

About the Company

Normal Computing builds silicon that turns thermal noise from an obstacle into a computational resource. We co-design the full stack: AI-native EDA systems and advanced ASICs that enable 10-100× more AI inference per dollar and per watt.

Responsibilities

• Develop algorithms for transformer inference workloads running on stochastic analog processing-with-memory hardware
• Work directly with hardware and architecture teams to shape chip capabilities through hardware co-design
• Design numerical methods that exploit thermal noise and analog dynamics
• Build evaluation frameworks and benchmarks to characterize algorithm behavior on real hardware or simulation
• Translate model workload insights into constraints and opportunities for hardware design
• Prototype and iterate rapidly as hardware evolves from simulation to silicon

Requirements

• Deep understanding of large model inference including attention mechanisms, KV cache, and long-context decoding
• Experience with inference optimization such as quantization, sparsity, kernel fusion, or memory-efficient attention
• Familiarity with stochastic systems, probabilistic methods, numerical analysis, or analog computation
• Experience implementing algorithms close to hardware
• Strong programming skills in Python and at least one systems language
• Ability to reason from first principles about novel substrates
• Track record of taking ideas from theory to working implementation on real hardware

Preferred Qualifications

• PhD in machine learning, applied mathematics, physics, electrical engineering, or a related field
• Exposure to analog or mixed-signal systems, in-memory compute, or non-von-Neumann architectures
• Experience working on hardware that did not yet exist when you joined
• Publications or open-source work in efficient inference, stochastic algorithms, or novel computing