ChipNexus.AI for AI-Native Computing

Modern hardware development means making challenging choices between simulators, synthesis engines, linters, verification environments, and build systems, often across different vendors and platforms. ChipsNexus.AI has developed NEX, a powerful, extensible automation framework that bridges the gap between hardware engineers and EDA tooling.

Students on this project will be installing NEX on the NSF Chipshub for dissemination in AI hardware courses. The project will demonstrate understanding that emerging AI workloads demand new hardware and system architectures, while AI is becoming a powerful tool for designing computing systems themselves.

Module 1 - Hardware for AI - Profile, serve, schedule, map, accelerate, and make reliable emerging LLM, agentic, physical, and compositional AI workloads

  • Explain the fundamentals of large language models (LLMs) and their applications in hardware design.
  • Develop prompt-based LLM frameworks for hardware design using prompt engineering, iterative feedback, and retrieval-augmented generation.
  • Fine-tune LLMs on domain-specific datasets for Verilog generation using supervised fine-tuning and reinforcement learning.

Module 2 - AI for Hardware - Use agents to design, optimize, and verify software, compilers, architectures, SoCs, RTL, EDA flows, and chips.

  • Develop LLM-based agents that complete multi-step hardware-design tasks, such as generating a processor design.
  • Construct, evaluate, and optimize multi-agent workflows for extended hardware-design tasks across the electronic design automation (EDA) flow.


 

Name of research group, project, or lab
Morrison VLSI and Spaceflight Lab
Logistics Information:
Project categories
Computer Science & Engineering
Electrical Engineering
Student ranks applicable
Junior
Senior
Student qualifications

A in Digital Integrated Circuits and previous experience with ChipNexus

Hours per week
2 credits / 6-12 hours
3 credits / 12+ hours
Compensation
Research for Credit
Number of openings
3
Techniques learned
  1. Co‑Pilot Mode
    • Interactive collaboration: Work side‑by‑side with NEX to explore ideas, investigate issues, and make progress in real time.
    • Intent‑driven tooling: Express goals and actions in natural language while NEX translates them into precise EDA and system commands.
    • Context‑aware reasoning & planning: Reason about design behavior, failures, and next steps using live context from waveforms, logs, and project state.
  2. Agent Mode
    • Autonomous workflow execution: Delegate complex, multi‑step tasks to NEX, from planning through execution and validation.
    • Self‑healing automation: When a build or simulation fails, NEX analyzes the results, adapts the workflow, and retries without manual intervention.
  3. Headless Mode
    • Scheduled, context‑aware work: Run regressions, synthesis, analysis, or reporting automatically on a schedule or trigger, with intelligent notifications on completion.
    • Background productivity: Continuously handle long‑running or routine tasks so you can stay focused on higher‑level design and decision‑making.
Contact Information:
Mentor
mmorri22@nd.edu
Associate Teaching Professor
Name of project director or principal investigator
Matthew Morrison
Email address of project director or principal investigator
matt.morrison@nd.edu
3 sp. | 0 appl.
Hours per week
2 credits / 6-12 hours (+1)
2 credits / 6-12 hours3 credits / 12+ hours
Project categories
Electrical Engineering (+1)
Computer Science & EngineeringElectrical Engineering