Neurophos · 14 hours ago
Tech Lead, Performance and Power Analysis
San Francisco
Full-time
Onsite
Lead/Staff
10+ years expNeurophos is a pioneering startup revolutionizing AI computation with an innovative optical computing platform. They are seeking a Tech Lead for Performance and Power Analysis to establish modeling infrastructure and drive performance characterization and power optimization for their advanced technology.
Artificial Intelligence (AI)Data CenterHardware
Responsibilities
Lead the PPA modeling team (4+ engineers) focused on performance and power analysis
Architect and implement performance models with discrete-event timing and cycle-accurate simulation
Develop power and energy modeling frameworks for optical engines, SRAM arrays, and digital logic
Define PPA analysis methodologies and establish modeling best practices
Drive performance optimization through bottleneck identification and architectural trade-off analysis
Collaborate with the architecture team on the performance characterization of novel compute blocks
Build and maintain a trace-driven simulation infrastructure for independent performance analysis
Develop power models for optical components, photonic devices, and opto-electronic interfaces
Work with silicon design teams to validate models against RTL and post-layout results
Mentor modeling engineers and establish team development practices
Present PPA results to the executive team and drive architecture decisions
Qualification
Required
MS or PhD in Computer Engineering, Electrical Engineering, or Computer Science (or BS with equivalent experience)
10+ years of experience in performance modeling and power analysis for CPUs, GPUs, or accelerators
Proven experience building and leading technical teams (3+ years of management experience)
Deep expertise in discrete-event simulation, cycle-accurate modeling, and performance analysis
Strong background in power modeling frameworks (McPAT, Cacti, or custom methodologies)
Expert-level C++ programming with a focus on performance-critical simulation code
Experience with trace-driven simulation and performance bottleneck analysis
Track record of shipping performance models that drove silicon design decisions
Excellent communication skills and ability to present technical results to diverse audiences
Experience with performance characterization of ML workloads on specialized hardware
Preferred
Experience with GPU performance modeling or shader core analysis
Background in accelerator architectures (TPU, NPU, DSP) or domain-specific processors
Knowledge of memory system modeling (HBM, DRAM controllers, cache hierarchies)
Familiarity with optical computing, photonics, or analog computing paradigms
Experience with event-driven simulation frameworks (SystemC, gem5, SST)
Understanding of ML framework internals (PyTorch, TensorFlow) and workload characteristics
Background in statistical performance modeling and regression analysis
Experience with power optimization techniques for datacenter processors
Publication record in computer architecture or hardware modeling venues
Benefits
Competitive compensation, including salary and equity options.
Opportunities for career growth and future team leadership.
Access to cutting-edge technology and state-of-the-art facilities.
Opportunity to publish research and contribute to the field of efficient AI inference.
Company
Neurophos
Photonic ExaOPS AI chips
11-50 employees
H1B Sponsorship
Neurophos has a track record of offering H1B sponsorships. Please note that this does not
guarantee sponsorship for this specific role. Below presents additional info for your
reference. (Data Powered by US Department of Labor)
Distribution of Different Job Fields Receiving Sponsorship
Represents job field similar to this job
Trends of Total Sponsorships
2023 (1)
Funding
Current Stage
Early StageTotal Funding
$16.1MKey Investors
Gates Frontier FundMetaVC Partners
2025-12-08Series Unknown· $1.9M
2023-12-14Seed· $7.2M
2023-10-03Seed· $7M
Leadership Team
Patrick T. Bowen
CEO
Andrew Traverso
Co-founder & Chief Scientist
Recent News
Trajectory Ventures
2026-01-21
2025-09-24
Company data provided by crunchbase