CPU Performance Modeling (Lead & ICs)
NuvacoreUnited States, Canada, India, RemoteRemote
Needs a free Superjobs account. You'll finish on Nuvacore's Lever page.
Skill score
Summary
The person owns Nuvacore's cycle-accurate CPU performance simulator and modeling methodology, leading microarchitectural exploration and model-to-RTL and post-silicon correlation. It suits a senior performance architect with deep C++ simulator experience who can also mentor engineers and lead technical direction.
- Function
- Development
- Estimated salary
- Est. $146k–220k/yr
- Estimated for a manager role, ~8 yrs experience, United States. Not published by the employer.
- Work mode and location
- Remote, United States, Canada, India, Remote
- Experience
- 8+ yrs
- Employment type
- Full-time
Job description
Roles and responsibilities
- Design, develop and own Nuvacore’s cycle-accurate CPU performance simulator — its architecture, fidelity, and long-term roadmap — as the authoritative reference model for the design team.
- Lead microarchitectural exploration across all major CPU domains: frontend (fetch, decode, branch prediction), out-of-order engine (rename, dispatch, execution), and memory subsystem (caches, prefetchers, TLBs, coherence).
- Drive model-to-RTL correlation through all execution phases; own debug and resolution of performance miscorrelation between the performance model, RTL simulation, and post-silicon results.
- Collaborate closely with design, verification, compiler, and system software teams to align architecture decisions with implementation constraints and software stack realities.
- Mentor engineers across the performance team; provide technical leadership on modeling methodology, simulation infrastructure, and analysis tooling.
- Contribute to IP development through patents and, where applicable, external publication
Skills required
Top skills
- Assembly programming
- CPU microarchitecture
- C++
- Python
All skills (8)
- Cycle-accurate CPU simulation
- gem5
- Branch prediction
- Out-of-order execution
- Memory subsystem
- RTL correlation
- RISC-V
- Perl
Qualifications
- MS in Computer Architecture, Computer Engineering, or related field (PhD preferred).
- 20+ years (Lead) or 8+ years (ICs) in CPU microarchitecture and/or performance engineering.
- 20+ years (Lead) or 8+ years (ICs) of hands-on experience with cycle-accurate C++ CPU performance simulators (e.g. gem5 or equivalent in-house tools).
- Deep expertise in at least one of: branch prediction, fetch/decode, rename/dispatch, OOO execution, memory subsystem, or prefetchers.
- Strong C++ and Python / Perl skills; ability to write and analyze assembly for microarchitectural test cases.
- Proven track record driving performance from pathfinding through silicon (RTL correlation and/or post-Si debug).
- Experience in RISC-V or ARM64 or x86 ISA.
- Strong track record in identifying ideas to improve power/performance/area (PPA) of CPUs.
- Experience developing trace-based and execution-driven CPU performance simulators.
Nice to have
- Experience with compiler optimizations and hardware/software co-design.
- Published work or filed patents in CPU performance or microarchitecture.
- Prior principal-level role at a CPU design organization.
Company info
About Nuvacore
At NUVACORE we're designing a new general-purpose CPU core for the sustained, intensive demands of AI infrastructure and agentic computing, not the workloads of the past. Founded by veteran CPU and system architects from world-class companies and backed by Sequoia Capital, we're a clean-sheet build with no legacy architecture to carry