Normalcomputing
Research Engineer, Algorithms
- Confirmed live in the last 24 hours
- $200k–$400k
- Mid level
- Full-time
- Remote · New York City
- Added 2 months ago
About this role
Normal Computing seeks a Research Engineer to develop algorithms specifically for their novel, stochastic analog hardware. This role involves co-designing algorithms and hardware, focusing on efficient transformer and diffusion model inference. The ideal candidate will blend a strong understanding of large model inference with expertise in stochastic systems and a proven ability to implement solutions on real hardware.
What you'll do
- Develop algorithms for transformer inference on stochastic analog hardware.
- Collaborate with hardware and architecture teams on chip design.
- Design numerical methods leveraging thermal noise and analog dynamics.
- Build evaluation frameworks and benchmarks for algorithm behavior.
- Translate model workload insights into hardware design constraints.
- Rapidly prototype and iterate as hardware evolves.
What they're looking for
- Large Model Inference
- Attention Mechanisms
- KV Cache
- Stochastic Systems
- Numerical Analysis
- Python
- Systems Language
- Algorithm Optimization
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Normalcomputing
Likely interview questions
- Describe your experience optimizing inference for large language models.
- Explain a time you had to work with hardware that was still under development.