Research Engineer, Algorithms

Job Title

Research Engineer, Algorithms

Role Summary

Develop algorithms and numerical methods to run AI inference efficiently on Normal's stochastic analog in-memory ASICs. Work in a hardware–software co-design team to map transformer and diffusion workloads onto the hardware's physical dynamics, validate on simulation and silicon, and influence architectural decisions.

Experience Level

Mid-level. Years of experience not specified; the role expects proven experience implementing inference systems on real hardware or high-fidelity simulation.

Responsibilities

Primary responsibilities include algorithm design, co-design with hardware teams, and building evaluation frameworks.

• Design algorithms for transformer and diffusion inference targeted to stochastic analog processing-in-memory.
• Develop numerical methods that exploit analog dynamics and thermal noise.
• Collaborate with hardware and architecture teams to shape chip capabilities and compute primitives.
• Build benchmarks and evaluation frameworks to characterize behavior on simulation and real silicon.
• Translate workload insights into hardware design constraints and opportunities.
• Rapidly prototype and iterate as hardware evolves from simulation to silicon.
• Optimize performance at both algorithmic and gate/implementation levels.

Requirements

Must-have technical skills and experience; nice-to-have items listed separately.

• Must-have: Deep understanding of large model inference (attention mechanisms, KV cache, long-context decoding, memory bandwidth constraints).
• Experience with inference optimization: quantization, sparsity, kernel fusion, or memory-efficient attention.
• Familiarity with stochastic systems, probabilistic methods, or numerical analysis as applied to computation.
• Experience implementing algorithms close to hardware (beyond high-level frameworks).
• Strong programming skills in Python and at least one systems language (e.g., C/C++, Rust).
• Proven track record of taking ideas from theory to working implementation on real hardware or high-fidelity simulation.
• Collaborative: able to work across hardware, architecture, and software teams.
• Nice-to-have: Exposure to analog or mixed-signal systems, in-memory compute, non-von-Neumann architectures; publications or open-source work in efficient inference or stochastic algorithms.

Education Requirements

PhD is listed as a bonus: "PhD in machine learning, applied mathematics, physics, electrical engineering, or a related field." No minimum degree is required and equivalent practical experience is implied but not explicitly stated.

About the Company

Company: Normal Computing

Headquarters: New York, NY, USA

Normal Computing develops software and hardware solutions for the semiconductor and AI infrastructure industries, specializing in ASICs for image and video diffusion inference and AI accelerators. The company focuses on architecture and microarchitecture of compute blocks, PE array design, ISA co-design, and FPGA prototyping.