AI Research Resident (Algorithms Codesign)

Normal Computing Corporation • New York City, NY

8d

About The Position

Normal Computing is hiring an AI Research Resident for its Algorithms Codesign track. The residency focuses on developing computational methods for efficient large-model inference on in-memory hardware, where computation takes place inside memory rather than moving weights to a processor. Residents work on a co-design problem, developing algorithms in parallel with the hardware so research findings can inform architectural decisions. Ideas are validated using simulators and FPGA prototypes, and later against silicon as it becomes available.

The resident will own a technical problem in inference algorithms for in-memory compute and work alongside researchers and engineers to turn research into production-quality research code. The program provides a dedicated research mentor and direct access to the team developing algorithms for in-memory inference hardware. Each residency is organized around two milestones: co-authoring a research paper with the team and presenting findings at a company-wide research colloquium. The role involves working in a fast-moving startup environment with ambiguity and rapid iteration.

Requirements

• Currently pursuing or recently completed a graduate degree (MS or PhD), or equivalent research experience, in machine learning, applied mathematics, physics, electrical engineering, computer science, or a related field.
• A solid research track record; publications are not required at the resident level.
• Working understanding of large-model inference, including attention mechanisms, KV cache, long-context decoding, and memory bandwidth constraints.
• Exposure to inference optimization (such as quantization, sparsity, kernel fusion, or memory-efficient attention), or to stochastic systems, probabilistic methods, numerical analysis, or analog computation.
• Strong Python skills.
• Some experience with a systems language such as C++ or Rust, or implementing algorithms close to hardware beyond high-level frameworks.
• Experience turning a research idea into working, reasonably reproducible code.
• Care about whether results replicate, not only whether a demo works once.
• Ability to reason from first principles about what a novel computing substrate can do efficiently.
• Strong written and verbal communication skills.
• Bias toward ownership and self-direction, and comfort with ambiguity and the pace of a small, fast-moving startup.

Nice To Haves

• Exposure to analog or mixed-signal systems, in-memory compute, or non-von-Neumann architectures.
• Publications or open-source work in efficient inference, stochastic algorithms, or novel computing.
• Experience working on hardware that did not yet exist when you joined.
• Prior experience writing research papers, theses, or technical reports.
• Experience presenting technical work to a live audience.

Responsibilities

• Develop inference algorithms for in-memory hardware, identifying LLM workload components suited to in-memory execution and demonstrating algorithms in simulation and on hardware.
• Design numerical methods that exploit thermal noise and analog dynamics.
• Build evaluation frameworks and benchmarks to characterize algorithm behavior in simulation and on real hardware.
• Translate insights about model workloads into constraints and opportunities for hardware and architecture teams.
• Prototype and iterate as the hardware evolves from simulation to FPGA to silicon.
• Partner with researchers and engineers to scope, conduct, and iterate on an original technical investigation, including rigorous experimental analysis and documentation.
• Co-author a research paper for submission to a relevant venue, with team mentorship on framing, experiments, and technical writing.
• Present work and findings at the residency colloquium and receive feedback from the Normal Computing research community.
• Contribute ideas and develop research into production-quality code that the team can build on.

What This Job Offers

• Job Type: Full-time
• Career Level: Intern