Research Engineer, Algorithms
At a glance
Mid-level AI Research role at Normal Computing Corporation. New York · full-time · $200,000–$400,000 base.
aimlroles summary, based on the employer's posting.
What you'll do
- Develop algorithms for transformer inference on hardware that performs stochastic analog computation within memory
- Shape chip architecture with hardware teams by identifying operations the system should handle natively
- Create numerical methods that use thermal noise and analog behavior in their computations
- Build benchmarks to test algorithm performance on real silicon or high-fidelity simulation
- Turn model workload findings into hardware design constraints, then refine prototypes as the hardware develops
What you bring
- You understand large-model inference, including attention, KV caching, long-context decoding and memory-bandwidth limits
- You have optimized inference using methods such as quantization, sparsity, kernel fusion or memory-efficient attention
- You bring knowledge of stochastic systems, probabilistic methods, numerical analysis or analog computation
- You have taken algorithms from theory to implementation on real hardware and can program in Python plus a systems language
Who this fits
This role suits an engineer who can reason from first principles and work across software, algorithms and hardware architecture. It is based in New York with a hybrid arrangement.
From the employer
Normal Computing | Build with Us
Normal is an applied AI company solving the hardest problems in AI and silicon. We build foundational hardware and software for the semiconductor industry, critical AI infrastructure, and the broader systems that power our world, in partnership with the world's most advanced institutions. We work as one team across New York City, Silicon Valley (Mountain View), London, Copenhagen, and Seoul.
The Role
You will develop the computational methods that make AI inference run efficiently on Normal's thermodynamic hardware. The core challenge is not adapting standard GPU kernels to a new chip. It is rethinking how operations like attention, memory access, and long-context decoding behave when the underlying substrate uses stochastic analog computation in memory rather than conventional digital logic.
Normal's ASICs run the heaviest operations of large model inference inside memory itself. Your job is to develop the algorithms that exploit this natively: understand what transformer and diffusion workloads are well-suited to stochastic analog execution, design numerical methods that map onto the hardware's physical dynamics, and validate them against real silicon or high-fidelity simulation.
This is a co-design role. The hardware and the algorithms are developed in parallel, which means you will influence architectural decisions, not just implement against a fixed specification. The strongest candidates have a deep understanding of both large model inference and the mathematics of stochastic systems, and have built systems that run on real hardware, not just in theory.
What You Will Own
Algorithm Development: Develop algorithms for transformer inference workloads running on stochastic analog processing-with-memory hardware.
Software/Hardware Co-Design: Work directly with hardware and architecture teams to shape what the chip can and should compute natively.
Numerical Methods: Design numerical methods that exploit thermal noise and analog dynamics rather than working around them.
Evaluation & Benchmarks: Build evaluation frameworks and benchmarks that characterize algorithm behavior on real hardware or simulation.
Workload Translation: Translate insights about model workloads into constraints and opportunities for hardware design.
Rapid Prototyping: Prototype and iterate rapidly as hardware evolves from simulation to silicon.
Optimizing Performance: at the gate level and the algorithmic level. and algorithms (expand/review), reinforcement learning tools.
What Makes You a Great Fit
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, numerical analysis, or analog computation
Experience implementing algorithms close to hardware, not just in high-level frameworks
Comfort reasoning from first principles about what a novel substrate can do efficiently
Track record of taking ideas from theory to working implementation on real hardware
Strong programming skills in Python and at least one systems language
Collaborative instinct and ability to work across hardware, architecture, and software teams
Bonus Points
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
Equal Employment Opportunity Statement
Normal Computing is an Equal Opportunity Employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, veteran status, or any other legally protected status.
Accessibility Accommodations
Normal Computing is committed to providing reasonable accommodations to individuals with disabilities. If you need assistance or an accommodation due to a disability, please let us know at accommodations@normalcomputing.com.
Privacy Notice
By submitting your application, you agree that Normal Computing may collect, use, and store your personal information for employment-related purposes in accordance with our Applicant Privacy Notice.